# Scene Flow to Action Map: A New Representation for RGB-D based Action Recognition with Convolutional Neural Networks

arXiv: https://arxiv.org/abs/1702.08652 · HTML: https://mlsanity.com/paper/1702.08652/

## Key facts

- Cited by 20 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 28 Feb 2017.

## Built on by (most cited first, 20 of 20)

- [Skeleton Cloud Colorization for Unsupervised 3D Action Representation Learning](https://mlsanity.com/paper/2108.01959/) (2021)
- [Trear: Transformer-based RGB-D Egocentric Action Recognition](https://mlsanity.com/paper/2101.03904/) (2021)
- [Contrastive Positive Mining for Unsupervised 3D Action Representation Learning](https://mlsanity.com/paper/2208.03497/) (2022)
- [milliFlow: Scene Flow Estimation on mmWave Radar Point Cloud for Human Motion Sensing](https://mlsanity.com/paper/2306.17010/) (2023)
- [Self-Supervised 3D Action Representation Learning with Skeleton Cloud Colorization](https://mlsanity.com/paper/2304.08799/) (2023)
- [A Comparative Analysis of Decision-Level Fusion for Multimodal Driver Behaviour Understanding](https://mlsanity.com/paper/2204.04734/) (2022)
- [M-FUSE: Multi-frame Fusion for Scene Flow Estimation](https://mlsanity.com/paper/2207.05704/) (2022)
- [Decoupling and Recoupling Spatiotemporal Representation for RGB-D-based Motion Recognition](https://mlsanity.com/paper/2112.09129/) (2021)
- [One-Shot Action Recognition via Multi-Scale Spatial-Temporal Skeleton Matching](https://mlsanity.com/paper/2307.07286/) (2023)
- [A Unified Multimodal De- and Re-coupling Framework for RGB-D Motion Recognition](https://mlsanity.com/paper/2211.09146/) (2022)
- [ModSelect: Automatic Modality Selection for Synthetic-to-Real Domain Generalization](https://mlsanity.com/paper/2208.09414/) (2022)
- [Deep video representation learning: a survey](https://mlsanity.com/paper/2405.06574/) (2024)
- [Transformers in Action Recognition: A Review on Temporal Modeling](https://mlsanity.com/paper/2302.01921/) (2022)
- [MVSA-Net: Multi-View State-Action Recognition for Robust and Deployable Trajectory Generation](https://mlsanity.com/paper/2311.08393/) (2023)
- [Cross-view Action Recognition via Contrastive View-invariant Representation](https://mlsanity.com/paper/2305.01733/) (2023)
- [Multi-stage Factorized Spatio-Temporal Representation for RGB-D Action and Gesture Recognition](https://mlsanity.com/paper/2308.12006/) (2023)
- [Egocentric RGB+Depth Action Recognition in Industry-Like Settings](https://mlsanity.com/paper/2309.13962/) (2023)
- [DeepMoCap: Deep Optical Motion Capture Using Multiple Depth Sensors and Retro-Reflectors](https://mlsanity.com/paper/2110.07283/) (2021)
- [Learning Scene Flow With Skeleton Guidance For 3D Action Recognition](https://mlsanity.com/paper/2306.13285/) (2023)
- [Skeleton-to-Image Encoding: Enabling Skeleton Representation Learning via Vision-Pretrained Models](https://mlsanity.com/paper/2603.05963/) (2026)

## Abstract (arXiv)

Scene flow describes the motion of 3D objects in real world and potentially could be the basis of a good feature for 3D action recognition. However, its use for action recognition, especially in the context of convolutional neural networks (ConvNets), has not been previously studied. In this paper, we propose the extraction and use of scene flow for action recognition from RGB-D data. Previous works have considered the depth and RGB modalities as separate channels and extract features for later fusion. We take a different approach and consider the modalities as one entity, thus allowing feature extraction for action recognition at the beginning. Two key questions about the use of scene flow for action recognition are addressed: how to organize the scene flow vectors and how to represent the long term dynamics of videos based on scene flow. In order to calculate the scene flow correctly on the available datasets, we propose an effective self-calibration method to align the RGB and depth data spatially without knowledge of the camera parameters. Based on the scene flow vectors, we propose a new representation, namely, Scene Flow to Action Map (SFAM), that describes several long term spatio-temporal dynamics for action recognition. We adopt a channel transform kernel to transform the scene flow vectors to an optimal color space analogous to RGB. This transformation takes better advantage of the trained ConvNets models over ImageNet. Experimental results indicate that this new representation can surpass the performance of state-of-the-art methods on two large public datasets.