Deep Video Generation, Prediction and Completion of Human Action Sequences

Reported on 4 benchmarks across 2 tasks · 1 paper · 4 SOTA

Note: results are matched by exact model name. Different papers may use the same name for different model variants.