Attentional Pooling for Action Recognition

Abstract

We introduce a simple yet surprisingly powerful model to incorporate attention in action recognition and human object interaction tasks. Our proposed attention module can be trained with or without extra supervision, and gives a sizable boost in accuracy while keeping the network size and computational cost nearly the same. It leads to significant improvements over state of the art base architecture on three standard action recognition benchmarks across still images and videos, and establishes new state of the art on MPII dataset with 12.5% relative improvement. We also perform an extensive analysis of our attention module both empirically and analytically. In terms of the latter, we introduce a novel derivation of bottom-up and top-down attention as low-rank approximations of bilinear pooling methods (typically used for fine-grained classification). From this perspective, our attention formulation suggests a novel characterization of action recognition as a fine-grained recognition problem.

In NIPS 2017. Project page: https://rohitgirdhar.github.io/AttentionalPoolingAction/

References in corpus (8)

• Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift
• UCF101: A Dataset of 101 Human Actions Classes From Videos in The Wild
• The Kinetics Human Action Video Dataset
• A simple neural network module for relational reasoning
• An End-to-End Spatio-Temporal Attention Model for Human Action Recognition from Skeleton Data
• Describing Common Human Visual Actions in Images
• Fine-grained Activity Recognition with Holistic and Pose based Features
• Joint Network based Attention for Action Recognition

Cited by in corpus (2)

• Learning Discriminative Motion Features Through Detection
• Loss Guided Activation for Action Recognition in Still Images