List your company
ResearchFormerly WizWand

Sign Language Recognition on Phoenix 2014T (test)

0.193WERTwoStream-SLR (Ours)
0.1728320.3089660.44510.581234Nov 2, 2022Feb 4, 2023May 10, 2023Aug 13, 2023Nov 15, 2023Feb 18, 2024May 23, 2024
Updated 8mo ago

Evaluation results

MethodLinks
TwoStream-SLR (Ours)Modality=RGB + Pose2022.110.193——
TwoStream-SLRModality=RGB-based, Pre-trained=Yes2024.050.193——
SignBERT+Modality=RGB2023.050.1994.33.9
SignBERT+ (+R)Modality=RGB-based, Pre-trained=Yes2024.050.199——
CTCAModality=RGB-based, Pre-trained=Yes2024.050.203——
C2SLR (RGB+Pose)Modality=RGB + Pose2022.110.204——
C2SLRModality=RGB-based, Pre-trained=Yes2024.050.204——
CorrNetModality=RGB-based, Pre-trained=Yes2024.050.205——
MSKA-SLRModality=Keypoint-based, Pre-trained=No2024.050.205——
AdaBrowse+Modality=RGB-based, Pre-trained=Yes2024.050.206——
STMC-R (RGB+Pose)Modality=RGB + Pose2022.110.21——
STMCModality=RGB2023.050.21——
STMCModality=RGB-based, Pre-trained=Yes2024.050.21——
SMKDModality=RGB2022.110.224——
MMTLBModality=RGB2022.110.225——
MMTLBModality=RGB-based, Pre-trained=Yes2024.050.225——
PiSLTRc-RModality=RGB2023.050.2295.14.4
DNFModality=RGB2023.050.2359.82.4
SignBTModality=RGB2022.110.239——
SignBTModality=RGB-based, Pre-trained=Yes2024.050.239——
CNN-LSTM-HMMs (Multi-Stream)Modality=RGB + hand/mouth shape2022.110.241——
3-streamModality=RGB2023.050.241——
CNN-LSTM-HMMModality=RGB-based, Pre-trained=No2024.050.241——
Joint-SLRTModality=RGB2022.110.245——
Joint-SLRTModality=RGB-based, Pre-trained=No2024.050.245——
SL-TransModality=RGB2023.050.24611.26.1
FCNModality=RGB2022.110.251——
FCNModality=RGB2023.050.251——
FCNModality=RGB-based, Pre-trained=No2024.050.251——
Self-Mutual KDExtraction Category=Gloss-based Feature Extraction2024.050.2538——
SFLModality=RGB2022.110.261——
1-streamModality=RGB2023.050.265——
TwoStream-SLRModality=Keypoint-based, Pre-trained=Yes2024.050.272——
Sign RecognitionExtraction Category=Gloss-based Feature Extraction2024.050.2959——
SignBERT+Modality=Pose2023.050.3368.45.3
SignBERT+Modality=Keypoint-based, Pre-trained=No2024.050.336——
P-TransModality=Pose2023.050.39811.43.8
P-BLSTMModality=Pose2023.050.40212.93.1
I3D PretrainedExtraction Category=Gloss-free Feature Extraction2024.050.6174——
SignCLExtraction Category=Gloss-free Feature Extraction, Base=VLP Pretrained2024.050.6333——
VLP PretrainedExtraction Category=Gloss-free Feature Extraction2024.050.6972——