Scalable frame resolution for efficient continuous sign language recognition

In this paper, we explore the spatial redundancy in continuous sign language recognition (CSLR), aiming to improve its efficiency. Despite recent advances in accuracy in CSLR, state-of-the-art CSLR methods typically require large amounts of computations and memory occupation, which are not friendly...
Ausführliche Beschreibung

Gespeichert in:

Autor*in:	Hu, Lianyu [verfasserIn] Gao, Liqing [verfasserIn] Liu, Zekang [verfasserIn] Feng, Wei [verfasserIn]

Format:	E-Artikel
Sprache:	Englisch

Erschienen:	2023

Schlagwörter:	Continuous sign language recognition Efficient inference Scalable frame resolution Adaptive inference

Übergeordnetes Werk:	Enthalten in: Pattern recognition - Amsterdam : Elsevier, 1968, 145
Übergeordnetes Werk:	volume:145

DOI / URN:	10.1016/j.patcog.2023.109903

Katalog-ID:	ELV06498298X

Internformat


LEADER	01000caa a22002652 4500
001	ELV06498298X
003	DE-627
005	20231205153554.0
007	cr uuu---uuuuu
008	231006s2023 xx \|\|\|\|\|o 00\| \|\|eng c
024	7		\|a 10.1016/j.patcog.2023.109903 \|2 doi
035			\|a (DE-627)ELV06498298X
035			\|a (ELSEVIER)S0031-3203(23)00601-5
040			\|a DE-627 \|b ger \|c DE-627 \|e rda
041			\|a eng
082	0	4	\|a 000 \|a 150 \|q VZ
084			\|a 54.74 \|2 bkl
100	1		\|a Hu, Lianyu \|e verfasserin \|0 (orcid)0000-0001-9974-4920 \|4 aut
245	1	0	\|a Scalable frame resolution for efficient continuous sign language recognition
264		1	\|c 2023
336			\|a nicht spezifiziert \|b zzz \|2 rdacontent
337			\|a Computermedien \|b c \|2 rdamedia
338			\|a Online-Ressource \|b cr \|2 rdacarrier
520			\|a In this paper, we explore the spatial redundancy in continuous sign language recognition (CSLR), aiming to improve its efficiency. Despite recent advances in accuracy in CSLR, state-of-the-art CSLR methods typically require large amounts of computations and memory occupation, which are not friendly towards fast inference under limited computation/memory budgets. Based on a simple observation that not all frames are equally important for CSLR, we propose AdaSize to handle this problem by modeling the frame resolution decision as an end-to-end learnable task to save unnecessary computations. Specifically, a lightweight 2D convolutional neural network (CNN) is first used to quickly browse input frames under a low resolution (e.g., 112 × 112). These extracted coarse and cheap features are sent into a recurrent policy network to dynamically determine the desired resolution for each frame. Once the optimal resolution for each frame is decided, frames with different resolutions are fed into the following backbones to extract representative features. Finally, these features pass through a sequence of temporal modules and a classifier to predict sentences. Extensive experiments on four large-scale datasets, including PHOENIX14, PHOENIX14-T, CSL-Daily and CSL, demonstrate the effectiveness of AdaSize. AdaSize could consistently achieve comparable accuracy with state-of-the-art CSLR methods, with only 0.38 × computations, 0.41 × memory usage and 1.25 × throughput. Comparisons with commonly-used lightweight backbones and other efficient methods verify the superiority of AdaSize under similar computational/memory budgets. We finally plot the frame resolution decisions for AdaSize, hoping to provide insightful analysis of the inherent spatial redundancy in videos.
650		4	\|a Continuous sign language recognition
650		4	\|a Efficient inference
650		4	\|a Scalable frame resolution
650		4	\|a Adaptive inference
700	1		\|a Gao, Liqing \|e verfasserin \|0 (orcid)0000-0003-4518-2154 \|4 aut
700	1		\|a Liu, Zekang \|e verfasserin \|4 aut
700	1		\|a Feng, Wei \|e verfasserin \|4 aut
773	0	8	\|i Enthalten in \|t Pattern recognition \|d Amsterdam : Elsevier, 1968 \|g 145 \|h Online-Ressource \|w (DE-627)265784131 \|w (DE-600)1466343-0 \|w (DE-576)101177364 \|7 nnns
773	1	8	\|g volume:145
912			\|a GBV_USEFLAG_U
912			\|a GBV_ELV
912			\|a SYSFLAG_U
912			\|a GBV_ILN_20
912			\|a GBV_ILN_22
912			\|a GBV_ILN_23
912			\|a GBV_ILN_24
912			\|a GBV_ILN_31
912			\|a GBV_ILN_32
912			\|a GBV_ILN_40
912			\|a GBV_ILN_60
912			\|a GBV_ILN_62
912			\|a GBV_ILN_65
912			\|a GBV_ILN_69
912			\|a GBV_ILN_70
912			\|a GBV_ILN_73
912			\|a GBV_ILN_74
912			\|a GBV_ILN_90
912			\|a GBV_ILN_95
912			\|a GBV_ILN_100
912			\|a GBV_ILN_101
912			\|a GBV_ILN_105
912			\|a GBV_ILN_110
912			\|a GBV_ILN_150
912			\|a GBV_ILN_151
912			\|a GBV_ILN_187
912			\|a GBV_ILN_213
912			\|a GBV_ILN_224
912			\|a GBV_ILN_230
912			\|a GBV_ILN_370
912			\|a GBV_ILN_602
912			\|a GBV_ILN_702
912			\|a GBV_ILN_2001
912			\|a GBV_ILN_2003
912			\|a GBV_ILN_2004
912			\|a GBV_ILN_2005
912			\|a GBV_ILN_2007
912			\|a GBV_ILN_2008
912			\|a GBV_ILN_2009
912			\|a GBV_ILN_2010
912			\|a GBV_ILN_2011
912			\|a GBV_ILN_2014
912			\|a GBV_ILN_2015
912			\|a GBV_ILN_2020
912			\|a GBV_ILN_2021
912			\|a GBV_ILN_2025
912			\|a GBV_ILN_2026
912			\|a GBV_ILN_2027
912			\|a GBV_ILN_2034
912			\|a GBV_ILN_2044
912			\|a GBV_ILN_2048
912			\|a GBV_ILN_2049
912			\|a GBV_ILN_2050
912			\|a GBV_ILN_2055
912			\|a GBV_ILN_2056
912			\|a GBV_ILN_2059
912			\|a GBV_ILN_2061
912			\|a GBV_ILN_2064
912			\|a GBV_ILN_2088
912			\|a GBV_ILN_2106
912			\|a GBV_ILN_2110
912			\|a GBV_ILN_2111
912			\|a GBV_ILN_2112
912			\|a GBV_ILN_2122
912			\|a GBV_ILN_2129
912			\|a GBV_ILN_2143
912			\|a GBV_ILN_2152
912			\|a GBV_ILN_2153
912			\|a GBV_ILN_2190
912			\|a GBV_ILN_2232
912			\|a GBV_ILN_2336
912			\|a GBV_ILN_2470
912			\|a GBV_ILN_2507
912			\|a GBV_ILN_4035
912			\|a GBV_ILN_4037
912			\|a GBV_ILN_4112
912			\|a GBV_ILN_4125
912			\|a GBV_ILN_4242
912			\|a GBV_ILN_4249
912			\|a GBV_ILN_4251
912			\|a GBV_ILN_4305
912			\|a GBV_ILN_4306
912			\|a GBV_ILN_4307
912			\|a GBV_ILN_4313
912			\|a GBV_ILN_4322
912			\|a GBV_ILN_4323
912			\|a GBV_ILN_4324
912			\|a GBV_ILN_4325
912			\|a GBV_ILN_4326
912			\|a GBV_ILN_4333
912			\|a GBV_ILN_4334
912			\|a GBV_ILN_4338
912			\|a GBV_ILN_4393
912			\|a GBV_ILN_4700
936	b	k	\|a 54.74 \|j Maschinelles Sehen \|q VZ
951			\|a AR
952			\|d 145

Indexfelder

author_variant	l h lh l g lg z l zl w f wf
matchkey_str	hulianyugaoliqingliuzekangfengwei:2023----:clbermrsltofrfiincniuusgl
hierarchy_sort_str	2023
bklnumber	54.74
publishDate	2023
allfields	10.1016/j.patcog.2023.109903 doi (DE-627)ELV06498298X (ELSEVIER)S0031-3203(23)00601-5 DE-627 ger DE-627 rda eng 000 150 VZ 54.74 bkl Hu, Lianyu verfasserin (orcid)0000-0001-9974-4920 aut Scalable frame resolution for efficient continuous sign language recognition 2023 nicht spezifiziert zzz rdacontent Computermedien c rdamedia Online-Ressource cr rdacarrier In this paper, we explore the spatial redundancy in continuous sign language recognition (CSLR), aiming to improve its efficiency. Despite recent advances in accuracy in CSLR, state-of-the-art CSLR methods typically require large amounts of computations and memory occupation, which are not friendly towards fast inference under limited computation/memory budgets. Based on a simple observation that not all frames are equally important for CSLR, we propose AdaSize to handle this problem by modeling the frame resolution decision as an end-to-end learnable task to save unnecessary computations. Specifically, a lightweight 2D convolutional neural network (CNN) is first used to quickly browse input frames under a low resolution (e.g., 112 × 112). These extracted coarse and cheap features are sent into a recurrent policy network to dynamically determine the desired resolution for each frame. Once the optimal resolution for each frame is decided, frames with different resolutions are fed into the following backbones to extract representative features. Finally, these features pass through a sequence of temporal modules and a classifier to predict sentences. Extensive experiments on four large-scale datasets, including PHOENIX14, PHOENIX14-T, CSL-Daily and CSL, demonstrate the effectiveness of AdaSize. AdaSize could consistently achieve comparable accuracy with state-of-the-art CSLR methods, with only 0.38 × computations, 0.41 × memory usage and 1.25 × throughput. Comparisons with commonly-used lightweight backbones and other efficient methods verify the superiority of AdaSize under similar computational/memory budgets. We finally plot the frame resolution decisions for AdaSize, hoping to provide insightful analysis of the inherent spatial redundancy in videos. Continuous sign language recognition Efficient inference Scalable frame resolution Adaptive inference Gao, Liqing verfasserin (orcid)0000-0003-4518-2154 aut Liu, Zekang verfasserin aut Feng, Wei verfasserin aut Enthalten in Pattern recognition Amsterdam : Elsevier, 1968 145 Online-Ressource (DE-627)265784131 (DE-600)1466343-0 (DE-576)101177364 nnns volume:145 GBV_USEFLAG_U GBV_ELV SYSFLAG_U GBV_ILN_20 GBV_ILN_22 GBV_ILN_23 GBV_ILN_24 GBV_ILN_31 GBV_ILN_32 GBV_ILN_40 GBV_ILN_60 GBV_ILN_62 GBV_ILN_65 GBV_ILN_69 GBV_ILN_70 GBV_ILN_73 GBV_ILN_74 GBV_ILN_90 GBV_ILN_95 GBV_ILN_100 GBV_ILN_101 GBV_ILN_105 GBV_ILN_110 GBV_ILN_150 GBV_ILN_151 GBV_ILN_187 GBV_ILN_213 GBV_ILN_224 GBV_ILN_230 GBV_ILN_370 GBV_ILN_602 GBV_ILN_702 GBV_ILN_2001 GBV_ILN_2003 GBV_ILN_2004 GBV_ILN_2005 GBV_ILN_2007 GBV_ILN_2008 GBV_ILN_2009 GBV_ILN_2010 GBV_ILN_2011 GBV_ILN_2014 GBV_ILN_2015 GBV_ILN_2020 GBV_ILN_2021 GBV_ILN_2025 GBV_ILN_2026 GBV_ILN_2027 GBV_ILN_2034 GBV_ILN_2044 GBV_ILN_2048 GBV_ILN_2049 GBV_ILN_2050 GBV_ILN_2055 GBV_ILN_2056 GBV_ILN_2059 GBV_ILN_2061 GBV_ILN_2064 GBV_ILN_2088 GBV_ILN_2106 GBV_ILN_2110 GBV_ILN_2111 GBV_ILN_2112 GBV_ILN_2122 GBV_ILN_2129 GBV_ILN_2143 GBV_ILN_2152 GBV_ILN_2153 GBV_ILN_2190 GBV_ILN_2232 GBV_ILN_2336 GBV_ILN_2470 GBV_ILN_2507 GBV_ILN_4035 GBV_ILN_4037 GBV_ILN_4112 GBV_ILN_4125 GBV_ILN_4242 GBV_ILN_4249 GBV_ILN_4251 GBV_ILN_4305 GBV_ILN_4306 GBV_ILN_4307 GBV_ILN_4313 GBV_ILN_4322 GBV_ILN_4323 GBV_ILN_4324 GBV_ILN_4325 GBV_ILN_4326 GBV_ILN_4333 GBV_ILN_4334 GBV_ILN_4338 GBV_ILN_4393 GBV_ILN_4700 54.74 Maschinelles Sehen VZ AR 145
spelling	10.1016/j.patcog.2023.109903 doi (DE-627)ELV06498298X (ELSEVIER)S0031-3203(23)00601-5 DE-627 ger DE-627 rda eng 000 150 VZ 54.74 bkl Hu, Lianyu verfasserin (orcid)0000-0001-9974-4920 aut Scalable frame resolution for efficient continuous sign language recognition 2023 nicht spezifiziert zzz rdacontent Computermedien c rdamedia Online-Ressource cr rdacarrier In this paper, we explore the spatial redundancy in continuous sign language recognition (CSLR), aiming to improve its efficiency. Despite recent advances in accuracy in CSLR, state-of-the-art CSLR methods typically require large amounts of computations and memory occupation, which are not friendly towards fast inference under limited computation/memory budgets. Based on a simple observation that not all frames are equally important for CSLR, we propose AdaSize to handle this problem by modeling the frame resolution decision as an end-to-end learnable task to save unnecessary computations. Specifically, a lightweight 2D convolutional neural network (CNN) is first used to quickly browse input frames under a low resolution (e.g., 112 × 112). These extracted coarse and cheap features are sent into a recurrent policy network to dynamically determine the desired resolution for each frame. Once the optimal resolution for each frame is decided, frames with different resolutions are fed into the following backbones to extract representative features. Finally, these features pass through a sequence of temporal modules and a classifier to predict sentences. Extensive experiments on four large-scale datasets, including PHOENIX14, PHOENIX14-T, CSL-Daily and CSL, demonstrate the effectiveness of AdaSize. AdaSize could consistently achieve comparable accuracy with state-of-the-art CSLR methods, with only 0.38 × computations, 0.41 × memory usage and 1.25 × throughput. Comparisons with commonly-used lightweight backbones and other efficient methods verify the superiority of AdaSize under similar computational/memory budgets. We finally plot the frame resolution decisions for AdaSize, hoping to provide insightful analysis of the inherent spatial redundancy in videos. Continuous sign language recognition Efficient inference Scalable frame resolution Adaptive inference Gao, Liqing verfasserin (orcid)0000-0003-4518-2154 aut Liu, Zekang verfasserin aut Feng, Wei verfasserin aut Enthalten in Pattern recognition Amsterdam : Elsevier, 1968 145 Online-Ressource (DE-627)265784131 (DE-600)1466343-0 (DE-576)101177364 nnns volume:145 GBV_USEFLAG_U GBV_ELV SYSFLAG_U GBV_ILN_20 GBV_ILN_22 GBV_ILN_23 GBV_ILN_24 GBV_ILN_31 GBV_ILN_32 GBV_ILN_40 GBV_ILN_60 GBV_ILN_62 GBV_ILN_65 GBV_ILN_69 GBV_ILN_70 GBV_ILN_73 GBV_ILN_74 GBV_ILN_90 GBV_ILN_95 GBV_ILN_100 GBV_ILN_101 GBV_ILN_105 GBV_ILN_110 GBV_ILN_150 GBV_ILN_151 GBV_ILN_187 GBV_ILN_213 GBV_ILN_224 GBV_ILN_230 GBV_ILN_370 GBV_ILN_602 GBV_ILN_702 GBV_ILN_2001 GBV_ILN_2003 GBV_ILN_2004 GBV_ILN_2005 GBV_ILN_2007 GBV_ILN_2008 GBV_ILN_2009 GBV_ILN_2010 GBV_ILN_2011 GBV_ILN_2014 GBV_ILN_2015 GBV_ILN_2020 GBV_ILN_2021 GBV_ILN_2025 GBV_ILN_2026 GBV_ILN_2027 GBV_ILN_2034 GBV_ILN_2044 GBV_ILN_2048 GBV_ILN_2049 GBV_ILN_2050 GBV_ILN_2055 GBV_ILN_2056 GBV_ILN_2059 GBV_ILN_2061 GBV_ILN_2064 GBV_ILN_2088 GBV_ILN_2106 GBV_ILN_2110 GBV_ILN_2111 GBV_ILN_2112 GBV_ILN_2122 GBV_ILN_2129 GBV_ILN_2143 GBV_ILN_2152 GBV_ILN_2153 GBV_ILN_2190 GBV_ILN_2232 GBV_ILN_2336 GBV_ILN_2470 GBV_ILN_2507 GBV_ILN_4035 GBV_ILN_4037 GBV_ILN_4112 GBV_ILN_4125 GBV_ILN_4242 GBV_ILN_4249 GBV_ILN_4251 GBV_ILN_4305 GBV_ILN_4306 GBV_ILN_4307 GBV_ILN_4313 GBV_ILN_4322 GBV_ILN_4323 GBV_ILN_4324 GBV_ILN_4325 GBV_ILN_4326 GBV_ILN_4333 GBV_ILN_4334 GBV_ILN_4338 GBV_ILN_4393 GBV_ILN_4700 54.74 Maschinelles Sehen VZ AR 145
allfields_unstemmed	10.1016/j.patcog.2023.109903 doi (DE-627)ELV06498298X (ELSEVIER)S0031-3203(23)00601-5 DE-627 ger DE-627 rda eng 000 150 VZ 54.74 bkl Hu, Lianyu verfasserin (orcid)0000-0001-9974-4920 aut Scalable frame resolution for efficient continuous sign language recognition 2023 nicht spezifiziert zzz rdacontent Computermedien c rdamedia Online-Ressource cr rdacarrier In this paper, we explore the spatial redundancy in continuous sign language recognition (CSLR), aiming to improve its efficiency. Despite recent advances in accuracy in CSLR, state-of-the-art CSLR methods typically require large amounts of computations and memory occupation, which are not friendly towards fast inference under limited computation/memory budgets. Based on a simple observation that not all frames are equally important for CSLR, we propose AdaSize to handle this problem by modeling the frame resolution decision as an end-to-end learnable task to save unnecessary computations. Specifically, a lightweight 2D convolutional neural network (CNN) is first used to quickly browse input frames under a low resolution (e.g., 112 × 112). These extracted coarse and cheap features are sent into a recurrent policy network to dynamically determine the desired resolution for each frame. Once the optimal resolution for each frame is decided, frames with different resolutions are fed into the following backbones to extract representative features. Finally, these features pass through a sequence of temporal modules and a classifier to predict sentences. Extensive experiments on four large-scale datasets, including PHOENIX14, PHOENIX14-T, CSL-Daily and CSL, demonstrate the effectiveness of AdaSize. AdaSize could consistently achieve comparable accuracy with state-of-the-art CSLR methods, with only 0.38 × computations, 0.41 × memory usage and 1.25 × throughput. Comparisons with commonly-used lightweight backbones and other efficient methods verify the superiority of AdaSize under similar computational/memory budgets. We finally plot the frame resolution decisions for AdaSize, hoping to provide insightful analysis of the inherent spatial redundancy in videos. Continuous sign language recognition Efficient inference Scalable frame resolution Adaptive inference Gao, Liqing verfasserin (orcid)0000-0003-4518-2154 aut Liu, Zekang verfasserin aut Feng, Wei verfasserin aut Enthalten in Pattern recognition Amsterdam : Elsevier, 1968 145 Online-Ressource (DE-627)265784131 (DE-600)1466343-0 (DE-576)101177364 nnns volume:145 GBV_USEFLAG_U GBV_ELV SYSFLAG_U GBV_ILN_20 GBV_ILN_22 GBV_ILN_23 GBV_ILN_24 GBV_ILN_31 GBV_ILN_32 GBV_ILN_40 GBV_ILN_60 GBV_ILN_62 GBV_ILN_65 GBV_ILN_69 GBV_ILN_70 GBV_ILN_73 GBV_ILN_74 GBV_ILN_90 GBV_ILN_95 GBV_ILN_100 GBV_ILN_101 GBV_ILN_105 GBV_ILN_110 GBV_ILN_150 GBV_ILN_151 GBV_ILN_187 GBV_ILN_213 GBV_ILN_224 GBV_ILN_230 GBV_ILN_370 GBV_ILN_602 GBV_ILN_702 GBV_ILN_2001 GBV_ILN_2003 GBV_ILN_2004 GBV_ILN_2005 GBV_ILN_2007 GBV_ILN_2008 GBV_ILN_2009 GBV_ILN_2010 GBV_ILN_2011 GBV_ILN_2014 GBV_ILN_2015 GBV_ILN_2020 GBV_ILN_2021 GBV_ILN_2025 GBV_ILN_2026 GBV_ILN_2027 GBV_ILN_2034 GBV_ILN_2044 GBV_ILN_2048 GBV_ILN_2049 GBV_ILN_2050 GBV_ILN_2055 GBV_ILN_2056 GBV_ILN_2059 GBV_ILN_2061 GBV_ILN_2064 GBV_ILN_2088 GBV_ILN_2106 GBV_ILN_2110 GBV_ILN_2111 GBV_ILN_2112 GBV_ILN_2122 GBV_ILN_2129 GBV_ILN_2143 GBV_ILN_2152 GBV_ILN_2153 GBV_ILN_2190 GBV_ILN_2232 GBV_ILN_2336 GBV_ILN_2470 GBV_ILN_2507 GBV_ILN_4035 GBV_ILN_4037 GBV_ILN_4112 GBV_ILN_4125 GBV_ILN_4242 GBV_ILN_4249 GBV_ILN_4251 GBV_ILN_4305 GBV_ILN_4306 GBV_ILN_4307 GBV_ILN_4313 GBV_ILN_4322 GBV_ILN_4323 GBV_ILN_4324 GBV_ILN_4325 GBV_ILN_4326 GBV_ILN_4333 GBV_ILN_4334 GBV_ILN_4338 GBV_ILN_4393 GBV_ILN_4700 54.74 Maschinelles Sehen VZ AR 145
allfieldsGer	10.1016/j.patcog.2023.109903 doi (DE-627)ELV06498298X (ELSEVIER)S0031-3203(23)00601-5 DE-627 ger DE-627 rda eng 000 150 VZ 54.74 bkl Hu, Lianyu verfasserin (orcid)0000-0001-9974-4920 aut Scalable frame resolution for efficient continuous sign language recognition 2023 nicht spezifiziert zzz rdacontent Computermedien c rdamedia Online-Ressource cr rdacarrier In this paper, we explore the spatial redundancy in continuous sign language recognition (CSLR), aiming to improve its efficiency. Despite recent advances in accuracy in CSLR, state-of-the-art CSLR methods typically require large amounts of computations and memory occupation, which are not friendly towards fast inference under limited computation/memory budgets. Based on a simple observation that not all frames are equally important for CSLR, we propose AdaSize to handle this problem by modeling the frame resolution decision as an end-to-end learnable task to save unnecessary computations. Specifically, a lightweight 2D convolutional neural network (CNN) is first used to quickly browse input frames under a low resolution (e.g., 112 × 112). These extracted coarse and cheap features are sent into a recurrent policy network to dynamically determine the desired resolution for each frame. Once the optimal resolution for each frame is decided, frames with different resolutions are fed into the following backbones to extract representative features. Finally, these features pass through a sequence of temporal modules and a classifier to predict sentences. Extensive experiments on four large-scale datasets, including PHOENIX14, PHOENIX14-T, CSL-Daily and CSL, demonstrate the effectiveness of AdaSize. AdaSize could consistently achieve comparable accuracy with state-of-the-art CSLR methods, with only 0.38 × computations, 0.41 × memory usage and 1.25 × throughput. Comparisons with commonly-used lightweight backbones and other efficient methods verify the superiority of AdaSize under similar computational/memory budgets. We finally plot the frame resolution decisions for AdaSize, hoping to provide insightful analysis of the inherent spatial redundancy in videos. Continuous sign language recognition Efficient inference Scalable frame resolution Adaptive inference Gao, Liqing verfasserin (orcid)0000-0003-4518-2154 aut Liu, Zekang verfasserin aut Feng, Wei verfasserin aut Enthalten in Pattern recognition Amsterdam : Elsevier, 1968 145 Online-Ressource (DE-627)265784131 (DE-600)1466343-0 (DE-576)101177364 nnns volume:145 GBV_USEFLAG_U GBV_ELV SYSFLAG_U GBV_ILN_20 GBV_ILN_22 GBV_ILN_23 GBV_ILN_24 GBV_ILN_31 GBV_ILN_32 GBV_ILN_40 GBV_ILN_60 GBV_ILN_62 GBV_ILN_65 GBV_ILN_69 GBV_ILN_70 GBV_ILN_73 GBV_ILN_74 GBV_ILN_90 GBV_ILN_95 GBV_ILN_100 GBV_ILN_101 GBV_ILN_105 GBV_ILN_110 GBV_ILN_150 GBV_ILN_151 GBV_ILN_187 GBV_ILN_213 GBV_ILN_224 GBV_ILN_230 GBV_ILN_370 GBV_ILN_602 GBV_ILN_702 GBV_ILN_2001 GBV_ILN_2003 GBV_ILN_2004 GBV_ILN_2005 GBV_ILN_2007 GBV_ILN_2008 GBV_ILN_2009 GBV_ILN_2010 GBV_ILN_2011 GBV_ILN_2014 GBV_ILN_2015 GBV_ILN_2020 GBV_ILN_2021 GBV_ILN_2025 GBV_ILN_2026 GBV_ILN_2027 GBV_ILN_2034 GBV_ILN_2044 GBV_ILN_2048 GBV_ILN_2049 GBV_ILN_2050 GBV_ILN_2055 GBV_ILN_2056 GBV_ILN_2059 GBV_ILN_2061 GBV_ILN_2064 GBV_ILN_2088 GBV_ILN_2106 GBV_ILN_2110 GBV_ILN_2111 GBV_ILN_2112 GBV_ILN_2122 GBV_ILN_2129 GBV_ILN_2143 GBV_ILN_2152 GBV_ILN_2153 GBV_ILN_2190 GBV_ILN_2232 GBV_ILN_2336 GBV_ILN_2470 GBV_ILN_2507 GBV_ILN_4035 GBV_ILN_4037 GBV_ILN_4112 GBV_ILN_4125 GBV_ILN_4242 GBV_ILN_4249 GBV_ILN_4251 GBV_ILN_4305 GBV_ILN_4306 GBV_ILN_4307 GBV_ILN_4313 GBV_ILN_4322 GBV_ILN_4323 GBV_ILN_4324 GBV_ILN_4325 GBV_ILN_4326 GBV_ILN_4333 GBV_ILN_4334 GBV_ILN_4338 GBV_ILN_4393 GBV_ILN_4700 54.74 Maschinelles Sehen VZ AR 145
allfieldsSound	10.1016/j.patcog.2023.109903 doi (DE-627)ELV06498298X (ELSEVIER)S0031-3203(23)00601-5 DE-627 ger DE-627 rda eng 000 150 VZ 54.74 bkl Hu, Lianyu verfasserin (orcid)0000-0001-9974-4920 aut Scalable frame resolution for efficient continuous sign language recognition 2023 nicht spezifiziert zzz rdacontent Computermedien c rdamedia Online-Ressource cr rdacarrier In this paper, we explore the spatial redundancy in continuous sign language recognition (CSLR), aiming to improve its efficiency. Despite recent advances in accuracy in CSLR, state-of-the-art CSLR methods typically require large amounts of computations and memory occupation, which are not friendly towards fast inference under limited computation/memory budgets. Based on a simple observation that not all frames are equally important for CSLR, we propose AdaSize to handle this problem by modeling the frame resolution decision as an end-to-end learnable task to save unnecessary computations. Specifically, a lightweight 2D convolutional neural network (CNN) is first used to quickly browse input frames under a low resolution (e.g., 112 × 112). These extracted coarse and cheap features are sent into a recurrent policy network to dynamically determine the desired resolution for each frame. Once the optimal resolution for each frame is decided, frames with different resolutions are fed into the following backbones to extract representative features. Finally, these features pass through a sequence of temporal modules and a classifier to predict sentences. Extensive experiments on four large-scale datasets, including PHOENIX14, PHOENIX14-T, CSL-Daily and CSL, demonstrate the effectiveness of AdaSize. AdaSize could consistently achieve comparable accuracy with state-of-the-art CSLR methods, with only 0.38 × computations, 0.41 × memory usage and 1.25 × throughput. Comparisons with commonly-used lightweight backbones and other efficient methods verify the superiority of AdaSize under similar computational/memory budgets. We finally plot the frame resolution decisions for AdaSize, hoping to provide insightful analysis of the inherent spatial redundancy in videos. Continuous sign language recognition Efficient inference Scalable frame resolution Adaptive inference Gao, Liqing verfasserin (orcid)0000-0003-4518-2154 aut Liu, Zekang verfasserin aut Feng, Wei verfasserin aut Enthalten in Pattern recognition Amsterdam : Elsevier, 1968 145 Online-Ressource (DE-627)265784131 (DE-600)1466343-0 (DE-576)101177364 nnns volume:145 GBV_USEFLAG_U GBV_ELV SYSFLAG_U GBV_ILN_20 GBV_ILN_22 GBV_ILN_23 GBV_ILN_24 GBV_ILN_31 GBV_ILN_32 GBV_ILN_40 GBV_ILN_60 GBV_ILN_62 GBV_ILN_65 GBV_ILN_69 GBV_ILN_70 GBV_ILN_73 GBV_ILN_74 GBV_ILN_90 GBV_ILN_95 GBV_ILN_100 GBV_ILN_101 GBV_ILN_105 GBV_ILN_110 GBV_ILN_150 GBV_ILN_151 GBV_ILN_187 GBV_ILN_213 GBV_ILN_224 GBV_ILN_230 GBV_ILN_370 GBV_ILN_602 GBV_ILN_702 GBV_ILN_2001 GBV_ILN_2003 GBV_ILN_2004 GBV_ILN_2005 GBV_ILN_2007 GBV_ILN_2008 GBV_ILN_2009 GBV_ILN_2010 GBV_ILN_2011 GBV_ILN_2014 GBV_ILN_2015 GBV_ILN_2020 GBV_ILN_2021 GBV_ILN_2025 GBV_ILN_2026 GBV_ILN_2027 GBV_ILN_2034 GBV_ILN_2044 GBV_ILN_2048 GBV_ILN_2049 GBV_ILN_2050 GBV_ILN_2055 GBV_ILN_2056 GBV_ILN_2059 GBV_ILN_2061 GBV_ILN_2064 GBV_ILN_2088 GBV_ILN_2106 GBV_ILN_2110 GBV_ILN_2111 GBV_ILN_2112 GBV_ILN_2122 GBV_ILN_2129 GBV_ILN_2143 GBV_ILN_2152 GBV_ILN_2153 GBV_ILN_2190 GBV_ILN_2232 GBV_ILN_2336 GBV_ILN_2470 GBV_ILN_2507 GBV_ILN_4035 GBV_ILN_4037 GBV_ILN_4112 GBV_ILN_4125 GBV_ILN_4242 GBV_ILN_4249 GBV_ILN_4251 GBV_ILN_4305 GBV_ILN_4306 GBV_ILN_4307 GBV_ILN_4313 GBV_ILN_4322 GBV_ILN_4323 GBV_ILN_4324 GBV_ILN_4325 GBV_ILN_4326 GBV_ILN_4333 GBV_ILN_4334 GBV_ILN_4338 GBV_ILN_4393 GBV_ILN_4700 54.74 Maschinelles Sehen VZ AR 145
language	English
source	Enthalten in Pattern recognition 145 volume:145
sourceStr	Enthalten in Pattern recognition 145 volume:145
format_phy_str_mv	Article
bklname	Maschinelles Sehen
institution	findex.gbv.de
topic_facet	Continuous sign language recognition Efficient inference Scalable frame resolution Adaptive inference
dewey-raw	000
isfreeaccess_bool	false
container_title	Pattern recognition
authorswithroles_txt_mv	Hu, Lianyu @@aut@@ Gao, Liqing @@aut@@ Liu, Zekang @@aut@@ Feng, Wei @@aut@@
publishDateDaySort_date	2023-01-01T00:00:00Z
hierarchy_top_id	265784131
dewey-sort	0
id	ELV06498298X
language_de	englisch
fullrecord	<?xml version="1.0" encoding="UTF-8"?><collection xmlns="http://www.loc.gov/MARC21/slim"><record><leader>01000caa a22002652 4500</leader><controlfield tag="001">ELV06498298X</controlfield><controlfield tag="003">DE-627</controlfield><controlfield tag="005">20231205153554.0</controlfield><controlfield tag="007">cr uuu---uuuuu</controlfield><controlfield tag="008">231006s2023 xx \|\|\|\|\|o 00\| \|\|eng c</controlfield><datafield tag="024" ind1="7" ind2=" "><subfield code="a">10.1016/j.patcog.2023.109903</subfield><subfield code="2">doi</subfield></datafield><datafield tag="035" ind1=" " ind2=" "><subfield code="a">(DE-627)ELV06498298X</subfield></datafield><datafield tag="035" ind1=" " ind2=" "><subfield code="a">(ELSEVIER)S0031-3203(23)00601-5</subfield></datafield><datafield tag="040" ind1=" " ind2=" "><subfield code="a">DE-627</subfield><subfield code="b">ger</subfield><subfield code="c">DE-627</subfield><subfield code="e">rda</subfield></datafield><datafield tag="041" ind1=" " ind2=" "><subfield code="a">eng</subfield></datafield><datafield tag="082" ind1="0" ind2="4"><subfield code="a">000</subfield><subfield code="a">150</subfield><subfield code="q">VZ</subfield></datafield><datafield tag="084" ind1=" " ind2=" "><subfield code="a">54.74</subfield><subfield code="2">bkl</subfield></datafield><datafield tag="100" ind1="1" ind2=" "><subfield code="a">Hu, Lianyu</subfield><subfield code="e">verfasserin</subfield><subfield code="0">(orcid)0000-0001-9974-4920</subfield><subfield code="4">aut</subfield></datafield><datafield tag="245" ind1="1" ind2="0"><subfield code="a">Scalable frame resolution for efficient continuous sign language recognition</subfield></datafield><datafield tag="264" ind1=" " ind2="1"><subfield code="c">2023</subfield></datafield><datafield tag="336" ind1=" " ind2=" "><subfield code="a">nicht spezifiziert</subfield><subfield code="b">zzz</subfield><subfield code="2">rdacontent</subfield></datafield><datafield tag="337" ind1=" " ind2=" "><subfield code="a">Computermedien</subfield><subfield code="b">c</subfield><subfield code="2">rdamedia</subfield></datafield><datafield tag="338" ind1=" " ind2=" "><subfield code="a">Online-Ressource</subfield><subfield code="b">cr</subfield><subfield code="2">rdacarrier</subfield></datafield><datafield tag="520" ind1=" " ind2=" "><subfield code="a">In this paper, we explore the spatial redundancy in continuous sign language recognition (CSLR), aiming to improve its efficiency. Despite recent advances in accuracy in CSLR, state-of-the-art CSLR methods typically require large amounts of computations and memory occupation, which are not friendly towards fast inference under limited computation/memory budgets. Based on a simple observation that not all frames are equally important for CSLR, we propose AdaSize to handle this problem by modeling the frame resolution decision as an end-to-end learnable task to save unnecessary computations. Specifically, a lightweight 2D convolutional neural network (CNN) is first used to quickly browse input frames under a low resolution (e.g., 112 × 112). These extracted coarse and cheap features are sent into a recurrent policy network to dynamically determine the desired resolution for each frame. Once the optimal resolution for each frame is decided, frames with different resolutions are fed into the following backbones to extract representative features. Finally, these features pass through a sequence of temporal modules and a classifier to predict sentences. Extensive experiments on four large-scale datasets, including PHOENIX14, PHOENIX14-T, CSL-Daily and CSL, demonstrate the effectiveness of AdaSize. AdaSize could consistently achieve comparable accuracy with state-of-the-art CSLR methods, with only 0.38 × computations, 0.41 × memory usage and 1.25 × throughput. Comparisons with commonly-used lightweight backbones and other efficient methods verify the superiority of AdaSize under similar computational/memory budgets. We finally plot the frame resolution decisions for AdaSize, hoping to provide insightful analysis of the inherent spatial redundancy in videos.</subfield></datafield><datafield tag="650" ind1=" " ind2="4"><subfield code="a">Continuous sign language recognition</subfield></datafield><datafield tag="650" ind1=" " ind2="4"><subfield code="a">Efficient inference</subfield></datafield><datafield tag="650" ind1=" " ind2="4"><subfield code="a">Scalable frame resolution</subfield></datafield><datafield tag="650" ind1=" " ind2="4"><subfield code="a">Adaptive inference</subfield></datafield><datafield tag="700" ind1="1" ind2=" "><subfield code="a">Gao, Liqing</subfield><subfield code="e">verfasserin</subfield><subfield code="0">(orcid)0000-0003-4518-2154</subfield><subfield code="4">aut</subfield></datafield><datafield tag="700" ind1="1" ind2=" "><subfield code="a">Liu, Zekang</subfield><subfield code="e">verfasserin</subfield><subfield code="4">aut</subfield></datafield><datafield tag="700" ind1="1" ind2=" "><subfield code="a">Feng, Wei</subfield><subfield code="e">verfasserin</subfield><subfield code="4">aut</subfield></datafield><datafield tag="773" ind1="0" ind2="8"><subfield code="i">Enthalten in</subfield><subfield code="t">Pattern recognition</subfield><subfield code="d">Amsterdam : Elsevier, 1968</subfield><subfield code="g">145</subfield><subfield code="h">Online-Ressource</subfield><subfield code="w">(DE-627)265784131</subfield><subfield code="w">(DE-600)1466343-0</subfield><subfield code="w">(DE-576)101177364</subfield><subfield code="7">nnns</subfield></datafield><datafield tag="773" ind1="1" ind2="8"><subfield code="g">volume:145</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_USEFLAG_U</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ELV</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">SYSFLAG_U</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_20</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_22</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_23</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_24</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_31</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_32</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_40</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_60</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_62</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_65</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_69</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_70</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_73</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_74</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_90</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_95</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_100</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_101</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_105</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_110</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_150</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_151</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_187</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_213</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_224</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_230</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_370</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_602</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_702</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_2001</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_2003</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_2004</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_2005</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_2007</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_2008</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_2009</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_2010</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_2011</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_2014</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_2015</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_2020</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_2021</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_2025</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_2026</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_2027</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_2034</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_2044</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_2048</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_2049</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_2050</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_2055</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_2056</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_2059</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_2061</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_2064</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_2088</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_2106</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_2110</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_2111</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_2112</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_2122</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_2129</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_2143</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_2152</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_2153</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_2190</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_2232</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_2336</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_2470</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_2507</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_4035</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_4037</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_4112</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_4125</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_4242</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_4249</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_4251</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_4305</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_4306</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_4307</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_4313</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_4322</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_4323</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_4324</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_4325</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_4326</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_4333</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_4334</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_4338</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_4393</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_4700</subfield></datafield><datafield tag="936" ind1="b" ind2="k"><subfield code="a">54.74</subfield><subfield code="j">Maschinelles Sehen</subfield><subfield code="q">VZ</subfield></datafield><datafield tag="951" ind1=" " ind2=" "><subfield code="a">AR</subfield></datafield><datafield tag="952" ind1=" " ind2=" "><subfield code="d">145</subfield></datafield></record></collection>
author	Hu, Lianyu
spellingShingle	Hu, Lianyu ddc 000 bkl 54.74 misc Continuous sign language recognition misc Efficient inference misc Scalable frame resolution misc Adaptive inference Scalable frame resolution for efficient continuous sign language recognition
authorStr	Hu, Lianyu
ppnlink_with_tag_str_mv	@@773@@(DE-627)265784131
format	electronic Article
dewey-ones	000 - Computer science, information & general works 150 - Psychology
delete_txt_mv	keep
author_role	aut aut aut aut
collection	elsevier
remote_str	true
illustrated	Not Illustrated
topic_title	000 150 VZ 54.74 bkl Scalable frame resolution for efficient continuous sign language recognition Continuous sign language recognition Efficient inference Scalable frame resolution Adaptive inference
topic	ddc 000 bkl 54.74 misc Continuous sign language recognition misc Efficient inference misc Scalable frame resolution misc Adaptive inference
topic_unstemmed	ddc 000 bkl 54.74 misc Continuous sign language recognition misc Efficient inference misc Scalable frame resolution misc Adaptive inference
topic_browse	ddc 000 bkl 54.74 misc Continuous sign language recognition misc Efficient inference misc Scalable frame resolution misc Adaptive inference
format_facet	Elektronische Aufsätze Aufsätze Elektronische Ressource
format_main_str_mv	Text Zeitschrift/Artikel
carriertype_str_mv	cr
hierarchy_parent_title	Pattern recognition
hierarchy_parent_id	265784131
dewey-tens	000 - Computer science, knowledge & systems 150 - Psychology
hierarchy_top_title	Pattern recognition
isfreeaccess_txt	false
familylinks_str_mv	(DE-627)265784131 (DE-600)1466343-0 (DE-576)101177364
title	Scalable frame resolution for efficient continuous sign language recognition
ctrlnum	(DE-627)ELV06498298X (ELSEVIER)S0031-3203(23)00601-5
title_full	Scalable frame resolution for efficient continuous sign language recognition
author_sort	Hu, Lianyu
journal	Pattern recognition
journalStr	Pattern recognition
lang_code	eng
isOA_bool	false
dewey-hundreds	000 - Computer science, information & general works 100 - Philosophy & psychology
recordtype	marc
publishDateSort	2023
contenttype_str_mv	zzz
author_browse	Hu, Lianyu Gao, Liqing Liu, Zekang Feng, Wei
container_volume	145
class	000 150 VZ 54.74 bkl
format_se	Elektronische Aufsätze
author-letter	Hu, Lianyu
doi_str_mv	10.1016/j.patcog.2023.109903
normlink	(ORCID)0000-0001-9974-4920 (ORCID)0000-0003-4518-2154
normlink_prefix_str_mv	(orcid)0000-0001-9974-4920 (orcid)0000-0003-4518-2154
dewey-full	000 150
author2-role	verfasserin
title_sort	scalable frame resolution for efficient continuous sign language recognition
title_auth	Scalable frame resolution for efficient continuous sign language recognition
abstract	In this paper, we explore the spatial redundancy in continuous sign language recognition (CSLR), aiming to improve its efficiency. Despite recent advances in accuracy in CSLR, state-of-the-art CSLR methods typically require large amounts of computations and memory occupation, which are not friendly towards fast inference under limited computation/memory budgets. Based on a simple observation that not all frames are equally important for CSLR, we propose AdaSize to handle this problem by modeling the frame resolution decision as an end-to-end learnable task to save unnecessary computations. Specifically, a lightweight 2D convolutional neural network (CNN) is first used to quickly browse input frames under a low resolution (e.g., 112 × 112). These extracted coarse and cheap features are sent into a recurrent policy network to dynamically determine the desired resolution for each frame. Once the optimal resolution for each frame is decided, frames with different resolutions are fed into the following backbones to extract representative features. Finally, these features pass through a sequence of temporal modules and a classifier to predict sentences. Extensive experiments on four large-scale datasets, including PHOENIX14, PHOENIX14-T, CSL-Daily and CSL, demonstrate the effectiveness of AdaSize. AdaSize could consistently achieve comparable accuracy with state-of-the-art CSLR methods, with only 0.38 × computations, 0.41 × memory usage and 1.25 × throughput. Comparisons with commonly-used lightweight backbones and other efficient methods verify the superiority of AdaSize under similar computational/memory budgets. We finally plot the frame resolution decisions for AdaSize, hoping to provide insightful analysis of the inherent spatial redundancy in videos.
abstractGer	In this paper, we explore the spatial redundancy in continuous sign language recognition (CSLR), aiming to improve its efficiency. Despite recent advances in accuracy in CSLR, state-of-the-art CSLR methods typically require large amounts of computations and memory occupation, which are not friendly towards fast inference under limited computation/memory budgets. Based on a simple observation that not all frames are equally important for CSLR, we propose AdaSize to handle this problem by modeling the frame resolution decision as an end-to-end learnable task to save unnecessary computations. Specifically, a lightweight 2D convolutional neural network (CNN) is first used to quickly browse input frames under a low resolution (e.g., 112 × 112). These extracted coarse and cheap features are sent into a recurrent policy network to dynamically determine the desired resolution for each frame. Once the optimal resolution for each frame is decided, frames with different resolutions are fed into the following backbones to extract representative features. Finally, these features pass through a sequence of temporal modules and a classifier to predict sentences. Extensive experiments on four large-scale datasets, including PHOENIX14, PHOENIX14-T, CSL-Daily and CSL, demonstrate the effectiveness of AdaSize. AdaSize could consistently achieve comparable accuracy with state-of-the-art CSLR methods, with only 0.38 × computations, 0.41 × memory usage and 1.25 × throughput. Comparisons with commonly-used lightweight backbones and other efficient methods verify the superiority of AdaSize under similar computational/memory budgets. We finally plot the frame resolution decisions for AdaSize, hoping to provide insightful analysis of the inherent spatial redundancy in videos.
abstract_unstemmed	In this paper, we explore the spatial redundancy in continuous sign language recognition (CSLR), aiming to improve its efficiency. Despite recent advances in accuracy in CSLR, state-of-the-art CSLR methods typically require large amounts of computations and memory occupation, which are not friendly towards fast inference under limited computation/memory budgets. Based on a simple observation that not all frames are equally important for CSLR, we propose AdaSize to handle this problem by modeling the frame resolution decision as an end-to-end learnable task to save unnecessary computations. Specifically, a lightweight 2D convolutional neural network (CNN) is first used to quickly browse input frames under a low resolution (e.g., 112 × 112). These extracted coarse and cheap features are sent into a recurrent policy network to dynamically determine the desired resolution for each frame. Once the optimal resolution for each frame is decided, frames with different resolutions are fed into the following backbones to extract representative features. Finally, these features pass through a sequence of temporal modules and a classifier to predict sentences. Extensive experiments on four large-scale datasets, including PHOENIX14, PHOENIX14-T, CSL-Daily and CSL, demonstrate the effectiveness of AdaSize. AdaSize could consistently achieve comparable accuracy with state-of-the-art CSLR methods, with only 0.38 × computations, 0.41 × memory usage and 1.25 × throughput. Comparisons with commonly-used lightweight backbones and other efficient methods verify the superiority of AdaSize under similar computational/memory budgets. We finally plot the frame resolution decisions for AdaSize, hoping to provide insightful analysis of the inherent spatial redundancy in videos.
collection_details	GBV_USEFLAG_U GBV_ELV SYSFLAG_U GBV_ILN_20 GBV_ILN_22 GBV_ILN_23 GBV_ILN_24 GBV_ILN_31 GBV_ILN_32 GBV_ILN_40 GBV_ILN_60 GBV_ILN_62 GBV_ILN_65 GBV_ILN_69 GBV_ILN_70 GBV_ILN_73 GBV_ILN_74 GBV_ILN_90 GBV_ILN_95 GBV_ILN_100 GBV_ILN_101 GBV_ILN_105 GBV_ILN_110 GBV_ILN_150 GBV_ILN_151 GBV_ILN_187 GBV_ILN_213 GBV_ILN_224 GBV_ILN_230 GBV_ILN_370 GBV_ILN_602 GBV_ILN_702 GBV_ILN_2001 GBV_ILN_2003 GBV_ILN_2004 GBV_ILN_2005 GBV_ILN_2007 GBV_ILN_2008 GBV_ILN_2009 GBV_ILN_2010 GBV_ILN_2011 GBV_ILN_2014 GBV_ILN_2015 GBV_ILN_2020 GBV_ILN_2021 GBV_ILN_2025 GBV_ILN_2026 GBV_ILN_2027 GBV_ILN_2034 GBV_ILN_2044 GBV_ILN_2048 GBV_ILN_2049 GBV_ILN_2050 GBV_ILN_2055 GBV_ILN_2056 GBV_ILN_2059 GBV_ILN_2061 GBV_ILN_2064 GBV_ILN_2088 GBV_ILN_2106 GBV_ILN_2110 GBV_ILN_2111 GBV_ILN_2112 GBV_ILN_2122 GBV_ILN_2129 GBV_ILN_2143 GBV_ILN_2152 GBV_ILN_2153 GBV_ILN_2190 GBV_ILN_2232 GBV_ILN_2336 GBV_ILN_2470 GBV_ILN_2507 GBV_ILN_4035 GBV_ILN_4037 GBV_ILN_4112 GBV_ILN_4125 GBV_ILN_4242 GBV_ILN_4249 GBV_ILN_4251 GBV_ILN_4305 GBV_ILN_4306 GBV_ILN_4307 GBV_ILN_4313 GBV_ILN_4322 GBV_ILN_4323 GBV_ILN_4324 GBV_ILN_4325 GBV_ILN_4326 GBV_ILN_4333 GBV_ILN_4334 GBV_ILN_4338 GBV_ILN_4393 GBV_ILN_4700
title_short	Scalable frame resolution for efficient continuous sign language recognition
remote_bool	true
author2	Gao, Liqing Liu, Zekang Feng, Wei
author2Str	Gao, Liqing Liu, Zekang Feng, Wei
ppnlink	265784131
mediatype_str_mv	c
isOA_txt	false
hochschulschrift_bool	false
doi_str	10.1016/j.patcog.2023.109903
up_date	2024-07-06T21:24:49.277Z
_version_	1803866434991816704
fullrecord_marcxml	<?xml version="1.0" encoding="UTF-8"?><collection xmlns="http://www.loc.gov/MARC21/slim"><record><leader>01000caa a22002652 4500</leader><controlfield tag="001">ELV06498298X</controlfield><controlfield tag="003">DE-627</controlfield><controlfield tag="005">20231205153554.0</controlfield><controlfield tag="007">cr uuu---uuuuu</controlfield><controlfield tag="008">231006s2023 xx \|\|\|\|\|o 00\| \|\|eng c</controlfield><datafield tag="024" ind1="7" ind2=" "><subfield code="a">10.1016/j.patcog.2023.109903</subfield><subfield code="2">doi</subfield></datafield><datafield tag="035" ind1=" " ind2=" "><subfield code="a">(DE-627)ELV06498298X</subfield></datafield><datafield tag="035" ind1=" " ind2=" "><subfield code="a">(ELSEVIER)S0031-3203(23)00601-5</subfield></datafield><datafield tag="040" ind1=" " ind2=" "><subfield code="a">DE-627</subfield><subfield code="b">ger</subfield><subfield code="c">DE-627</subfield><subfield code="e">rda</subfield></datafield><datafield tag="041" ind1=" " ind2=" "><subfield code="a">eng</subfield></datafield><datafield tag="082" ind1="0" ind2="4"><subfield code="a">000</subfield><subfield code="a">150</subfield><subfield code="q">VZ</subfield></datafield><datafield tag="084" ind1=" " ind2=" "><subfield code="a">54.74</subfield><subfield code="2">bkl</subfield></datafield><datafield tag="100" ind1="1" ind2=" "><subfield code="a">Hu, Lianyu</subfield><subfield code="e">verfasserin</subfield><subfield code="0">(orcid)0000-0001-9974-4920</subfield><subfield code="4">aut</subfield></datafield><datafield tag="245" ind1="1" ind2="0"><subfield code="a">Scalable frame resolution for efficient continuous sign language recognition</subfield></datafield><datafield tag="264" ind1=" " ind2="1"><subfield code="c">2023</subfield></datafield><datafield tag="336" ind1=" " ind2=" "><subfield code="a">nicht spezifiziert</subfield><subfield code="b">zzz</subfield><subfield code="2">rdacontent</subfield></datafield><datafield tag="337" ind1=" " ind2=" "><subfield code="a">Computermedien</subfield><subfield code="b">c</subfield><subfield code="2">rdamedia</subfield></datafield><datafield tag="338" ind1=" " ind2=" "><subfield code="a">Online-Ressource</subfield><subfield code="b">cr</subfield><subfield code="2">rdacarrier</subfield></datafield><datafield tag="520" ind1=" " ind2=" "><subfield code="a">In this paper, we explore the spatial redundancy in continuous sign language recognition (CSLR), aiming to improve its efficiency. Despite recent advances in accuracy in CSLR, state-of-the-art CSLR methods typically require large amounts of computations and memory occupation, which are not friendly towards fast inference under limited computation/memory budgets. Based on a simple observation that not all frames are equally important for CSLR, we propose AdaSize to handle this problem by modeling the frame resolution decision as an end-to-end learnable task to save unnecessary computations. Specifically, a lightweight 2D convolutional neural network (CNN) is first used to quickly browse input frames under a low resolution (e.g., 112 × 112). These extracted coarse and cheap features are sent into a recurrent policy network to dynamically determine the desired resolution for each frame. Once the optimal resolution for each frame is decided, frames with different resolutions are fed into the following backbones to extract representative features. Finally, these features pass through a sequence of temporal modules and a classifier to predict sentences. Extensive experiments on four large-scale datasets, including PHOENIX14, PHOENIX14-T, CSL-Daily and CSL, demonstrate the effectiveness of AdaSize. AdaSize could consistently achieve comparable accuracy with state-of-the-art CSLR methods, with only 0.38 × computations, 0.41 × memory usage and 1.25 × throughput. Comparisons with commonly-used lightweight backbones and other efficient methods verify the superiority of AdaSize under similar computational/memory budgets. We finally plot the frame resolution decisions for AdaSize, hoping to provide insightful analysis of the inherent spatial redundancy in videos.</subfield></datafield><datafield tag="650" ind1=" " ind2="4"><subfield code="a">Continuous sign language recognition</subfield></datafield><datafield tag="650" ind1=" " ind2="4"><subfield code="a">Efficient inference</subfield></datafield><datafield tag="650" ind1=" " ind2="4"><subfield code="a">Scalable frame resolution</subfield></datafield><datafield tag="650" ind1=" " ind2="4"><subfield code="a">Adaptive inference</subfield></datafield><datafield tag="700" ind1="1" ind2=" "><subfield code="a">Gao, Liqing</subfield><subfield code="e">verfasserin</subfield><subfield code="0">(orcid)0000-0003-4518-2154</subfield><subfield code="4">aut</subfield></datafield><datafield tag="700" ind1="1" ind2=" "><subfield code="a">Liu, Zekang</subfield><subfield code="e">verfasserin</subfield><subfield code="4">aut</subfield></datafield><datafield tag="700" ind1="1" ind2=" "><subfield code="a">Feng, Wei</subfield><subfield code="e">verfasserin</subfield><subfield code="4">aut</subfield></datafield><datafield tag="773" ind1="0" ind2="8"><subfield code="i">Enthalten in</subfield><subfield code="t">Pattern recognition</subfield><subfield code="d">Amsterdam : Elsevier, 1968</subfield><subfield code="g">145</subfield><subfield code="h">Online-Ressource</subfield><subfield code="w">(DE-627)265784131</subfield><subfield code="w">(DE-600)1466343-0</subfield><subfield code="w">(DE-576)101177364</subfield><subfield code="7">nnns</subfield></datafield><datafield tag="773" ind1="1" ind2="8"><subfield code="g">volume:145</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_USEFLAG_U</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ELV</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">SYSFLAG_U</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_20</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_22</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_23</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_24</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_31</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_32</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_40</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_60</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_62</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_65</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_69</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_70</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_73</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_74</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_90</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_95</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_100</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_101</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_105</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_110</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_150</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_151</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_187</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_213</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_224</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_230</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_370</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_602</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_702</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_2001</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_2003</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_2004</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_2005</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_2007</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_2008</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_2009</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_2010</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_2011</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_2014</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_2015</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_2020</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_2021</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_2025</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_2026</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_2027</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_2034</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_2044</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_2048</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_2049</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_2050</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_2055</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_2056</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_2059</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_2061</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_2064</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_2088</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_2106</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_2110</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_2111</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_2112</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_2122</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_2129</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_2143</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_2152</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_2153</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_2190</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_2232</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_2336</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_2470</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_2507</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_4035</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_4037</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_4112</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_4125</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_4242</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_4249</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_4251</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_4305</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_4306</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_4307</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_4313</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_4322</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_4323</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_4324</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_4325</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_4326</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_4333</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_4334</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_4338</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_4393</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_4700</subfield></datafield><datafield tag="936" ind1="b" ind2="k"><subfield code="a">54.74</subfield><subfield code="j">Maschinelles Sehen</subfield><subfield code="q">VZ</subfield></datafield><datafield tag="951" ind1=" " ind2=" "><subfield code="a">AR</subfield></datafield><datafield tag="952" ind1=" " ind2=" "><subfield code="d">145</subfield></datafield></record></collection>
score	7.400216

Nicht das Richtige dabei?

Schreiben Sie uns!

Scalable frame resolution for efficient continuous sign language recognition

Nicht das Richtige dabei?

Zugang & Verfügbarkeit

Vorhandene Bände

Nicht das Richtige dabei?