Learning to Segment Human Body Parts with Synthetically Trained Deep Convolutional Networks

Abstract

This paper presents a new framework for human body part segmentation based on Deep Convolutional Neural Networks trained using only synthetic data. The proposed approach achieves cutting-edge results without the need of training the models with real annotated data of human body parts. Our contributions include a data generation pipeline, that exploits a game engine for the creation of the synthetic data used for training the network, and a novel pre-processing module, that combines edge response maps and adaptive histogram equalization to guide the network to learn the shape of the human body parts ensuring robustness to changes in the illumination conditions. For selecting the best candidate architecture, we perform exhaustive tests on manually annotated images of real human body limbs. We further compare our method against several high-end commercial segmentation tools on the body parts segmentation task. The results show that our method outperforms the other models by a significant margin. Finally, we present an ablation study to validate our pre-processing module. With this paper, we release an implementation of the proposed approach along with the acquired datasets.

This paper has been published in: Proceedings of the 16th International Conference on Intelligent Autonomous Systems (IAS 2021)

References in corpus (8)

• Very Deep Convolutional Networks for Large-Scale Image Recognition
• Rethinking Atrous Convolution for Semantic Image Segmentation
• Hierarchical Multi-Scale Attention for Semantic Segmentation
• Building an Aerial-Ground Robotics System for Precision Farming: An Adaptable Solution
• Dual Path Networks
• Using CNNs For Users Segmentation In Video See-Through Augmented Virtuality
• Ego2Hands: A Dataset for Egocentric Two-hand Segmentation and Detection
• Enhanced Self-Perception in Mixed Reality: Egocentric Arm Segmentation and Database with Automatic Labelling