Learning Collision-Free Space Detection from Stereo Images: Homography Matrix Brings Better Data Augmentation

Abstract

Collision-free space detection is a critical component of autonomous vehicle perception. The state-of-the-art algorithms are typically based on supervised learning. The performance of such approaches is always dependent on the quality and amount of labeled training data. Additionally, it remains an open challenge to train deep convolutional neural networks (DCNNs) using only a small quantity of training samples. Therefore, this paper mainly explores an effective training data augmentation approach that can be employed to improve the overall DCNN performance, when additional images captured from different views are available. Due to the fact that the pixels of the collision-free space (generally regarded as a planar surface) between two images captured from different views can be associated by a homography matrix, the scenario of the target image can be transformed into the reference view. This provides a simple but effective way of generating training data from additional multi-view images. Extensive experimental results, conducted with six state-of-the-art semantic segmentation DCNNs on three datasets, demonstrate the effectiveness of our proposed training data augmentation algorithm for enhancing collision-free space detection performance. When validated on the KITTI road benchmark, our approach provides the best results for stereo vision-based collision-free space detection.

arXiv commentaccepted to IEEE/ASME Transactions on Mechatronics

References in corpus (13)

• Rethinking Atrous Convolution for Semantic Image Segmentation
• Encoder-Decoder with Atrous Separable Convolution for Semantic Image Segmentation
• Semantic Image Segmentation with Deep Convolutional Nets and Fully Connected CRFs
• A Survey of Deep Learning Techniques for Autonomous Driving
• Pothole Detection Based on Disparity Transformation and Road Surface Modeling
• SNE-RoadSeg: Incorporating Surface Normal Information into Semantic Segmentation for Accurate Freespace Detection
• Road surface 3d reconstruction based on dense subpixel disparity map estimation
• Road Damage Detection Based on Unsupervised Disparity Map Segmentation
• Dynamic Fusion Module Evolves Drivable Area and Road Anomaly Detection: A Benchmark and Algorithms
• Applying Surface Normal Information in Drivable Area and Road Anomaly Detection for Ground Mobile Robots
• We Learn Better Road Pothole Detection: from Attention Aggregation to Adversarial Domain Adaptation
• Real-time stereo vision-based lane detection system
• Computer Stereo Vision for Autonomous Driving

Cited by in corpus (10)

• Computer Vision for Road Imaging and Pothole Detection: A State-of-the-Art Review of Systems and Algorithms
• Dynamic Fusion Module Evolves Drivable Area and Road Anomaly Detection: A Benchmark and Algorithms
• PVStereo: Pyramid Voting Module for End-to-End Self-Supervised Stereo Matching
• Computer Stereo Vision for Autonomous Driving
• Rethinking Road Surface 3D Reconstruction and Pothole Detection: From Perspective Transformation to Disparity Map Segmentation
• CodingHomo: Bootstrapping Deep Homography With Video Coding
• Learning Interpretable End-to-End Vision-Based Motion Planning for Autonomous Driving with Optical Flow Distillation
• Co-Teaching: An Ark to Unsupervised Stereo Matching
• S2P2: Self-Supervised Goal-Directed Path Planning Using RGB-D Data for Robotic Wheelchairs
• SCV-Stereo: Learning Stereo Matching from a Sparse Cost Volume