Social Scene Understanding: End-to-End Multi-Person Action Localization and Collective Activity Recognition
This work addresses the challenge of automated social scene analysis for applications like surveillance or human-computer interaction, though it appears incremental as it builds on existing multi-person action localization methods.
The paper tackles the problem of understanding human social behaviors in image sequences by jointly detecting individuals, inferring their actions, and estimating collective activities in a single feed-forward pass, achieving state-of-the-art performance on multiple benchmarks.
We present a unified framework for understanding human social behaviors in raw image sequences. Our model jointly detects multiple individuals, infers their social actions, and estimates the collective actions with a single feed-forward pass through a neural network. We propose a single architecture that does not rely on external detection algorithms but rather is trained end-to-end to generate dense proposal maps that are refined via a novel inference scheme. The temporal consistency is handled via a person-level matching Recurrent Neural Network. The complete model takes as input a sequence of frames and outputs detections along with the estimates of individual actions and collective activities. We demonstrate state-of-the-art performance of our algorithm on multiple publicly available benchmarks.