LSTA

Reported on 2 benchmarks across 1 task · 1 paper · 2 SOTA

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

Robots2 results

Activity RecognitiononEPIC-KITCHENS-55
Actions Top-1 (S2)· 2018-11-26
16.63
best: 27.3 (DEEP-HAL with ODF+SDF (AssembleNet++))
SOTA
LSTA: Long Short-Term Attention for Egocentric Action Recognition arXiv:1811.10698
Activity RecognitiononEGTEA
Average Accuracy· 2018-11-26
61.9
best: 81.75 (LaViLa (Finetuned, TimeSformer-L))
SOTA
LSTA: Long Short-Term Attention for Egocentric Action Recognition arXiv:1811.10698