FBK-HUPBA Submission to the EPIC-Kitchens Action Recognition 2020 Challenge
This is an incremental improvement for the EPIC-Kitchens benchmark, addressing action recognition in egocentric kitchen videos.
The paper tackled action recognition in the EPIC-Kitchens 2020 Challenge by deploying spatio-temporal models like GSM and EgoACO in an ensemble, achieving top-1 accuracies of 40.0% on S1 and 25.71% on S2 using only RGB.
In this report we describe the technical details of our submission to the EPIC-Kitchens Action Recognition 2020 Challenge. To participate in the challenge we deployed spatio-temporal feature extraction and aggregation models we have developed recently: Gate-Shift Module (GSM) [1] and EgoACO, an extension of Long Short-Term Attention (LSTA) [2]. We design an ensemble of GSM and EgoACO model families with different backbones and pre-training to generate the prediction scores. Our submission, visible on the public leaderboard with team name FBK-HUPBA, achieved a top-1 action recognition accuracy of 40.0% on S1 setting, and 25.71% on S2 setting, using only RGB.