PERF-Net: Pose Empowered RGB-Flow Net
This work addresses the problem of improving action recognition accuracy for video analysis, though it is incremental as it builds on existing two-stream models.
The paper tackles video action recognition by adding a human pose stream to standard RGB and flow streams, showing that this simple addition provides complementary gains and leads to a model (PERF-Net) that outperforms state-of-the-art by a large margin on multiple datasets, without requiring explicit flow or pose computation at inference.
In recent years, many works in the video action recognition literature have shown that two stream models (combining spatial and temporal input streams) are necessary for achieving state of the art performance. In this paper we show the benefits of including yet another stream based on human pose estimated from each frame -- specifically by rendering pose on input RGB frames. At first blush, this additional stream may seem redundant given that human pose is fully determined by RGB pixel values -- however we show (perhaps surprisingly) that this simple and flexible addition can provide complementary gains. Using this insight, we then propose a new model, which we dub PERF-Net (short for Pose Empowered RGB-Flow Net), which combines this new pose stream with the standard RGB and flow based input streams via distillation techniques and show that our model outperforms the state-of-the-art by a large margin in a number of human action recognition datasets while not requiring flow or pose to be explicitly computed at inference time. The proposed pose stream is also part of the winner solution of the ActivityNet Kinetics Challenge 2020.