Temporal-Relational CrossTransformers for Few-Shot Action Recognition

We propose a novel approach to few-shot action recognition, finding temporally-corresponding frame tuples between the query and videos in the support set. Distinct from previous few-shot works, we construct class prototypes using the CrossTransformer attention mechanism to observe relevant sub-sequences of all support videos, rather than using class averages or single best matches. Video representations are formed from ordered tuples of varying numbers of frames, which allows sub-sequences of actions at different speeds and temporal offsets to be compared. Our proposed Temporal-Relational CrossTransformers (TRX) achieve state-of-the-art results on few-shot splits of Kinetics, Something-Something V2 (SSv2), HMDB51 and UCF101. Importantly, our method outperforms prior work on SSv2 by a wide margin (12%) due to the its ability to model temporal relations. A detailed ablation showcases the importance of matching to multiple support set videos and learning higher-order relational CrossTransformers.

PDF Abstract CVPR 2021 PDF CVPR 2021 Abstract
Task Dataset Model Metric Name Metric Value Global Rank Result Benchmark
Few Shot Action Recognition HMDB51 TRX 1:1 Accuracy 75.6 # 4
Few Shot Action Recognition Kinetics-100 TRX Accuracy 85.9 # 4
Few Shot Action Recognition Something-Something-100 TRX 1:1 Accuracy 64.6 # 3
Few Shot Action Recognition UCF101 TRX 1:1 Accuracy 96.1 # 3