List your company
ResearchFormerly WizWand

Continuous Sign Language Recognition on PHOENIX14-T (dev)

17WERDCA
16.36420.65724.9529.243Dec 26, 2022Jul 10, 2023Jan 22, 2024Aug 5, 2024Feb 17, 2025Sep 1, 2025Mar 17, 2026
Updated 7mo ago

Evaluation results

MethodLinks
DCAextra cues (keypoints)=false2023.0517
CorrNet+Venue=arXiv 20242025.1217.2
C2STextra cues (keypoints)=false2023.0517.3
USTMbaseModality=RGB2025.1217.4
USTMsmallModality=RGB2025.1217.4
USTMsmallBackbone=Swin-small2025.1217.4
USTMsmallVariant=small2025.1217.4
USTMbaseModality=RGB2025.1217.6
ACMCVenue=IMAVIS 20252025.1217.6
USTMbaseBackbone=Swin-base2025.1217.6
USTMbaseVariant=base2025.1217.6
TwoStream-SLRextra cues (keypoints)=true2023.0517.7
TwoStream-SLRModality=RGB+Pose2025.1217.7
SlowFastSignVenue=ICASSP 20242025.1217.7
USTMtinyModality=RGB2025.1217.8
USTMtinyBackbone=Swin-tiny2025.1217.8
USTMtinyVariant=tiny2025.1217.8
USTMsmallModality=RGB2025.1217.9
SlowFastSign (Reproduced)Venue=ICASSP 2024, reproducibility=Reproduced2025.1217.9
USTMtinyModality=RGB2025.1218.1
TwoStream-SLRModality=RGB+Pose2025.1218.4
Swin-MSTPVenue=Neurocomp. 2025, Backbone=Swin-small2025.1218.7
SignBERT+extra cues (keypoints)=false2023.0518.8
CorrNetextra cues (keypoints)=false2023.0518.9
CorrNetVenue=CVPR 20232025.1218.9
STNetVenue=IEEE Acc. 20252025.1219.2
SENBackbone=ResNet18, GFLOPS per video=3672023.0819.3
AdaBrowsealpha=0.04, GFLOPS=287, Throughput (videos/s)=12.312023.0819.3
SENextra cues (keypoints)=false2023.0519.3
CTCAextra cues (keypoints)=false2023.0519.3
SENVenue=AAAI 20232025.1219.3
CTCAVenue=CVPR 20232025.1219.3
TLPBackbone=ResNet18, GFLOPS per video=3672023.0819.4
AdaBrowseBackbone=ResNet18, GFLOPS per video=2542023.0819.4
AdaBrowsealpha=0.15, GFLOPS=254, Throughput (videos/s)=14.262023.0819.4
AdaBrowse+alpha=0.05, GFLOPS=241, Throughput (videos/s)=15.122023.0819.4
TLPextra cues (keypoints)=false2023.0519.4
CVT-SLRextra cues (keypoints)=false2023.0519.4
CVT-SLRModality=RGB+Pose2025.1219.4
TLPVenue=ECCV 20222025.1219.4
AdaBrowse+Backbone=ResNet18, GFLOPS per video=1712023.0819.5
BaselineGFLOPS=364, Throughput (videos/s)=12.222023.0819.5
AdaBrowse+alpha=0.1, GFLOPS=171, Throughput (videos/s)=17.582023.0819.5
CoSignextra cues (keypoints)=false2023.0519.5
AdaBrowseVenue=ACM MM 20232025.1219.5
CoSignPre-Training=false, Encoder Params=≈ 10M, Decoder Params=≈ 18M2026.0319.5
STMCBackbone=VGG11, GFLOPS per video=1554, Extra Modalities=Face and Hands2023.0819.6
STMCEnd-to-end=false, Modalities (Training)=R+P, Modalities (Inference)=R2022.1219.6
STMCModality=RGB+Pose2025.1219.6
CVT-SLRModality=RGB+Pose2025.1219.8
CLIP-SLAVenue=CVPR 20252025.1219.8
SignBERTModality=RGB+Pose2025.1220.1
MP-SLRVenue=TPAMI 20252025.1220.1
MSKAVenue=PR 20252025.1220.1
MSKAPre-Training=false, Encoder Params=≈ 15M, Decoder Params=≈ 28M2026.0320.1
C2SLRBackbone=ResNet18, GFLOPS per video=367, Extra Modalities=Face and Hands2023.0820.2
C2SLREnd-to-end=true, Modalities (Training)=R+P, Modalities (Inference)=R2022.1220.2
C2SLRextra cues (keypoints)=true2023.0520.2
C2SLRModality=RGB+Pose2025.1220.2
C2SLRModality=RGB+Pose2025.1220.5
SSSLRVenue=ICASSP 20232025.1220.5
STMCModality=RGB+Pose2025.1220.6
SMKDBackbone=ResNet18, GFLOPS per video=3672023.0820.8
SMKDEnd-to-end=false, Modalities (Training)=R, Modalities (Inference)=R2022.1220.8
SMKDVenue=ICCV 20212025.1220.8
STARKPre-Training=false, Encoder Params=≈ 3M, Decoder Params=≈ 28M2026.0321
MMTLBEnd-to-end=true, Modalities (Training)=R, Modalities (Inference)=R2022.1221.9
V-L Mapperextra cues (keypoints)=false2023.0521.9
CNN+HMM+LSTMBackbone=GoogLeNet, GFLOPS per video=316, Extra Modalities=Face and Hands2023.0822.1
CNN-LSTM-HMMsEnd-to-end=false, Modalities (Training)=R, Modalities (Inference)=R2022.1222.1
DNFModality=RGB+Op.Flow2025.1223.1
FCNBackbone=Custom, GFLOPS per video=2862023.0823.3
FCNEnd-to-end=true, Modalities (Training)=R, Modalities (Inference)=R2022.1223.3
FCNVenue=ECCV 20202025.1223.3
CMAVenue=ACM MM 20202025.1224.1
SLTBackbone=GoogLeNet, GFLOPS per video=316, Extra Modalities=Face and Hands2023.0824.5
SLTEnd-to-end=true, Modalities (Training)=R, Modalities (Inference)=R2022.1224.6
SFLEnd-to-end=true, Modalities (Training)=R, Modalities (Inference)=R2022.1225.1
TwoStream-SLRPre-Training=true2026.0327.1
SignBERT+Pre-Training=true2026.0332.9