Our new X account is live! Follow @wizwand_team for updates
WorkDL logo mark

TaiChi

Benchmarks

Task NameDataset NameSOTA ResultTrend
Video GenerationTaichi 256x256 (test)
Sampling Time (s)4.8
22
Landmark detectionTaichi (test)
L2 Distance316.1
8
Video GenerationTaichi-128x128
FVD (16 frames)94.6
6
Video generationTaiChi-256
FVD (16 frames)89.5
6
Part SegmentationTaichi
MAE389.78
4
Human motion transferTaiChi (OOD)
FID15.89
3
Video GenerationTaiChi
FVD128.1
2
Controlled Video SynthesisTaiChi 68 (test)
FVD152.76
2
Showing 8 of 8 rows