List your company

Video Semantic Segmentation on Cityscapes (val)

83.3mIoUMask2Former
45.44455.27265.174.928Apr 2, 2018Aug 3, 2019Dec 4, 2020Apr 7, 2022Aug 9, 2023Dec 10, 2024Apr 13, 2026

Evaluation results

MethodLinks
Mask2FormerBackbone=Swin-B2023.1083.3——1,0571073.26—
Mask2FormerBackbone=Swin-L2023.1083.3——1,911.32152.11—
Mask2FormerBackbone=Swin-S2023.1082.6——730.168.74.31—
SegFormerBackbone=MiT-B52023.1082.4——1,460.484.77.2—
Mask2FormerBackbone=Swin-T2023.1082.1——543.647.45.41—
MPVSSBackbone=Swin-B2023.1081.7——278.61479.54—
MPVSSBackbone=Swin-L2023.1081.6——449.5255.47.24—
Mask2Former-AccelBackbone=Swin-B+Swin-T2023.1081.4——680.1193.14.4—
MPVSSBackbone=Swin-S2023.1081.3——213.210810.98—
MPVSSBackbone=Swin-T2023.1080.7——175.911412.33—
NetwarpSpeed (fps)=0.33, Avg RT (ms)=3004, MaxLatency (ms)=30042020.0780.6——————
TMANet-50 (Ours-50)GFLOPs=7542021.0280.3——————
baselineDescription=per-frame segmentation, Backbone=ResNet-1012018.0480.2360360————
Mask2FormerBackbone=R1012023.1080.1——685.5635.68—
TD2-PSP50Speed (ms/f)=178, Max Latency (ms)=1782020.0479.9——————
TDNet-50GFLOPs=10822021.0279.9——————
Mask2Former-DFFBackbone=Swin-B2023.1079.9——525.3145.76.09—
PSPNet101Speed (ms/f)=360, Max Latency (ms)=3602020.0479.7——————
Mask2FormerBackbone=R502023.1079.4——529.9446.58—
Mask2Former-AccelBackbone=R101+R502023.1078.9——594.8145.75.78—
PSPNetBackbone=R1012023.1078.5——2,048.967.92.88—
SegFormerBackbone=MiT-B12023.1078.5——243.713.820.7—
MPVSSBackbone=R502023.1078.4——173.284.113.43—
RS-SSMBackbone=MiT-B1, Input size=512 x 10242026.0378.3——59.520.127.2—
MPVSSBackbone=R1012023.1078.2——204.3103.112.55—
PSPNet50Speed (ms/f)=238, Max Latency (ms)=2382020.0478.1——————
Mask2Former-DFFBackbone=R1012023.1077.1——457.4101.77.14—
AFP + AKSFeature Propagation Method=Adaptive (AFP), Key-frame Selection Strategy=Adaptive (AKS), Scheduling Scheme=standard, Backbone=ResNet-1012018.0476.84171380————
LVSSpeed (ms/f)=171, Max Latency (ms)=3802020.0476.8——————
TD4-PSP18Speed (ms/f)=85, Max Latency (ms)=852020.0476.8——————
LVS2021.0276.8——————
FANet34+TempSpeed (fps)=58, Avg RT (ms)=17, MaxLatency (ms)=17, Temporal Attention=t=22020.0776.7——————
FCNBackbone=R1012023.1076.6——2,203.368.52.83—
PEARLSpeed (ms/f)=800, Max Latency (ms)=8002020.0476.5——————
BiseNet* 101Speed (ms/f)=722020.0476.5——————
BiseNet* 101Speed (ms/f)=722020.0476.5——————
TD2-Bise34Speed (ms/f)=262020.0476.4——————
TD2-Bise34Speed (ms/f)=262020.0476.4——————
FANet34Speed (fps)=58, Avg RT (ms)=17, MaxLatency (ms)=172020.0776.3——————
BiseNet* 34Speed (ms/f)=272020.0476——————
BiseNet* 34Speed (ms/f)=272020.0476——————
LVS-LLSSpeed (ms/f)=119, Max Latency (ms)=1192020.0475.9——————
Low-LatencySpeed (fps)=7.5, Avg RT (ms)=133, MaxLatency (ms)=1332020.0775.9——————
AFP + AKS + LLSFeature Propagation Method=Adaptive (AFP), Key-frame Selection Strategy=Adaptive (AKS), Scheduling Scheme=Low-Latency (LLS), Backbone=ResNet-1012018.0475.89119119————
CFFM++Backbone=MiT-B1, Input size=512 x 10242026.0375.7——169.815.920.4—
TV3SBackbone=MiT-B1, Input size=512 x 10242026.0375.6——83.617.325.1—
PSPNet18Speed (ms/f)=91, Max Latency (ms)=912020.0475.5——————
FANet18+TempSpeed (fps)=72, Avg RT (ms)=14, MaxLatency (ms)=14, Temporal Attention=t=22020.0775.5——————
SwiftNetSpeed (ms/f)=232020.0475.4——————
SwiftNetSpeed (ms/f)=232020.0475.4——————
AFP + fix scheduleFeature Propagation Method=Adaptive (AFP), Key-frame Selection Strategy=fix schedule (l=5), Scheduling Scheme=standard, Backbone=ResNet-1012018.0475.26151360————
MRCFABackbone=MiT-B1, Params (M)=14.9, FPS (f/s)=21.52022.0775.1——————
CFFM-VSSBackbone=MiT-B12023.1075.1——158.715.411.71—
MRCFABackbone=MiT-B12023.1075.1——14514.912.97—
MRCFABackbone=MiT-B1, Input size=512 x 10242026.0375.1——14514.913—
TD4-Bise18Speed (ms/f)=212020.0475——————
TD4-Bise18Speed (ms/f)=212020.0475——————
FANet18Speed (fps)=72, Avg RT (ms)=14, MaxLatency (ms)=142020.0775——————
SegFormerBackbone=MiT-B1, Params (M)=13.8, FPS (f/s)=46.82022.0774.1——————
SegFormerBackbone=MiT-B1, Input size=512 x 10242026.0374.1——106.413.746.8—
CFFM-VSSBackbone=MiT-B02023.1074——80.74.615.79—
CFFMBackbone=MiT-B0, Input size=512 x 10242026.0374——80.74.615.8—
BiseNet* 18Speed (ms/f)=202020.0473.8——————
BiseNet* 18Speed (ms/f)=202020.0473.8——————
GRFP(5)Speed (ms/f)=255, Max Latency (ms)=2552020.0473.6——————
GRFP2021.0273.6——————
LadderNetSpeed (ms/f)=332020.0472.8——————
LadderNetSpeed (ms/f)=332020.0472.8——————
MRCFABackbone=MiT-B0, Params (M)=4.2, FPS (f/s)=33.32022.0772.8——————
MRCFABackbone=MiT-B02023.1072.8——77.54.216.55—
MRCFABackbone=MiT-B0, Input size=512 x 10242026.0372.8——77.54.216.6—
AccelSpeed (fps)=2.9, Avg RT (ms)=340, MaxLatency (ms)=5752020.0772.1——————
AccelBackbone=ResNet-101, FPS (f/s)=3.62022.0772.1——————
AccelBackbone=R1012023.1072.1——824.4———
SegFormerBackbone=MiT-B0, Params (M)=3.7, FPS (f/s)=58.52022.0771.9——————
SegFormerBackbone=MiT-B0, Input size=512 x 10242026.0371.9——54.63.758.5—
ETCBackbone=ResNet-18, Params (M)=13.2, FPS (f/s)=9.52022.0771.1——————
ETCBackbone=R182023.1071.1——434.1———
DVSNBackbone=ResNet-101, FPS (f/s)=19.82022.0770.3——————
DVSNBackbone=R1012023.1070.3——978.4———
PSPNetBackbone=MobileNetV2, Params (M)=13.7, FPS (f/s)=11.22022.0770.2——————
Deep Fea. Flow2018.0470.1273654————
GRFP(5)interval between keyframes=52018.0469.4470470————
GRFPBackbone=ResNet-101, FPS (f/s)=3.22022.0769.4——————
DFFSpeed (ms/f)=156, Max Latency (ms)=5752020.0469.2——————
DFFSpeed (fps)=5.7, Avg RT (ms)=175, MaxLatency (ms)=6442020.0769.2——————
DFFGFLOPs=>9192021.0269.2——————
DFFBackbone=ResNet-101, FPS (f/s)=9.72022.0768.7——————
DFFBackbone=R1012023.1068.7——100.8———
Clockwork NetBackbone=ResNet-101, interval between keyframes=52018.0467.7141360————
ICNetSpeed (ms/f)=202020.0467.7——————
ICNetSpeed (ms/f)=202020.0467.7——————
CCBackbone=VGG-16, FPS (f/s)=16.52022.0767.7——————
CLKSpeed (ms/f)=158, Max Latency (ms)=1982020.0464.4——————
ClockworkSpeed (fps)=5.6, Avg RT (ms)=177, MaxLatency (ms)=2212020.0764.4——————
DVSNetSpeed (ms/f)=332020.0463.2——————
DVSNetSpeed (ms/f)=332020.0463.2——————
DVSNet-fastSpeed (fps)=30.4, Avg RT (ms)=332020.0763.2——————
FCNBackbone=MobileNetV2, Params (M)=9.8, FPS (f/s)=14.22022.0761.5——————
DiTTAPre-trained on=VSPW2026.0446.9—————77.9