LoCATe-GAT: Modeling Multi-Scale Local Context and Action Relationships for Zero-Shot Action Recognition

Open paper
Benchmark Model Rank Results
zero-shot-action-recognition-on-activitynetLoCATe-GAT#3Top-1 Accuracy: 73.8
zero-shot-action-recognition-on-hmdb51LoCATe-GAT#7Top-1 Accuracy: 50.7
zero-shot-action-recognition-on-kineticsLoCATe-GAT#9Top-1 Accuracy: 58.7
zero-shot-action-recognition-on-ucf101LoCATe-GAT#9Top-1 Accuracy: 76.0

Rank counts only results with a code link.