FID (CS): 27.480FID (CV): 31.875
Papers with Code · CC BY-SA 4.0
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."
<span style="color:grey; opacity: 0.6">( Image credit: Convolutional Sequence Generation for Skeleton-Based Action Synthesis )</span>
논문
FID (CS): 3.618FID (CV): 4.235
Accuracy: 93.51FID: 34Diversity: 6.56Multimodality: 2.49
MMDa (CS): 0.698MMDs (CS): 0.788MMDa (CV): 0.999MMDs (CV): 1.311
MMDa (CS): 0.338MMDs (CS): 0.402MMDa (CV): 0.371MMDs (CV): 0.398
MMDa (CS): 0.285MMDs (CS): 0.299MMDa (CV): 0.316MMDs (CV): 0.335
FID (CS): 6.030FID (CV): 7.114