SOTAVerified

Human action generation

Yan et al. (2019) CSGN:

"When the dancer is stepping, jumping and spinning on the stage, attentions of all audiences are attracted by the streamof the fluent and graceful movements. Building a model that is capable of dancing is as fascinating a task as appreciating the performance itself. In this paper, we aim to generate long-duration human actions represented as skeleton sequences, e.g. those that cover the entirety of a dance, with hundreds of moves and countless possible combinations."

( Image credit: Convolutional Sequence Generation for Skeleton-Based Action Synthesis )

Papers

Showing 1113 of 13 papers

TitleStatusHype
Convolutional Sequence Generation for Skeleton-Based Action Synthesis0
Deep Video Generation, Prediction and Completion of Human Action Sequences0
FinePhys: Fine-grained Human Action Generation by Explicitly Incorporating Physical Laws for Effective Skeletal Guidance0
Show:102550
← PrevPage 2 of 2Next →

Benchmark Results

#ModelMetricClaimedVerifiedStatus
1Deep Video Generation, Prediction and Completion of Human Action SequencesMMDa0.42Unverified
2Learning Diverse Stochastic Human-Action Generators by Learning Smooth Latent TransitionsMMDa0.2Unverified
3c-GANMMDa0.16Unverified
4SA-GCNMMDa0.15Unverified
5Kinetic-GANMMDa0.07Unverified
#ModelMetricClaimedVerifiedStatus
1SkeletonGANMMDa (CS)0.7Unverified
2c-SkeletonGANMMDa (CS)0.34Unverified
3c-GANMMDa (CS)0.33Unverified
4SA-GCNMMDa (CS)0.29Unverified
5Kinetic-GANMMDa (CS)0.26Unverified
#ModelMetricClaimedVerifiedStatus
1c-GANFID (CS)27.48Unverified
2CSGNFID (CS)6.03Unverified
3Kinetic-GANFID (CS)3.62Unverified
#ModelMetricClaimedVerifiedStatus
1c-GANFID (CS)54.4Unverified
2Kinetic-GANFID (CS)5.97Unverified
#ModelMetricClaimedVerifiedStatus
1ODMOAccuracy93.51Unverified
#ModelMetricClaimedVerifiedStatus
1ODMOAccuracy97.81Unverified
#ModelMetricClaimedVerifiedStatus
1ODMOAccuracy93.67Unverified