SINet: A Scale-insensitive Convolutional Neural Network for Fast Vehicle Detection

Abstract

Vision-based vehicle detection approaches achieve incredible success in recent years with the development of deep convolutional neural network (CNN). However, existing CNN based algorithms suffer from the problem that the convolutional features are scale-sensitive in object detection task but it is common that traffic images and videos contain vehicles with a large variance of scales. In this paper, we delve into the source of scale sensitivity, and reveal two key issues: 1) existing RoI pooling destroys the structure of small scale objects, 2) the large intra-class distance for a large variance of scales exceeds the representation capability of a single network. Based on these findings, we present a scale-insensitive convolutional neural network (SINet) for fast detecting vehicles with a large variance of scales. First, we present a context-aware RoI pooling to maintain the contextual information and original structure of small scale objects. Second, we present a multi-branch decision network to minimize the intra-class distance of features. These lightweight techniques bring zero extra time complexity but prominent detection accuracy improvement. The proposed techniques can be equipped with any deep network architectures and keep them trained end-to-end. Our SINet achieves state-of-the-art performance in terms of accuracy and speed (up to 37 FPS) on the KITTI benchmark and a new highway dataset, which contains a large variance of scales and extremely small objects.

Accepted by IEEE Transactions on Intelligent Transportation Systems (T-ITS)

References in corpus (15)

• Very Deep Convolutional Networks for Large-Scale Image Recognition
• Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks
• How transferable are features in deep neural networks?
• R-FCN: Object Detection via Region-based Fully Convolutional Networks
• Spatial Pyramid Pooling in Deep Convolutional Networks for Visual Recognition
• Feature Pyramid Networks for Object Detection
• Aggregated Residual Transformations for Deep Neural Networks
• Beyond Skip Connections: Top-Down Modulation for Object Detection
• Embedding Structured Contour and Location Prior in Siamesed Fully Convolutional Networks for Road Detection
• PVANET: Deep but Lightweight Neural Networks for Real-time Object Detection
• A Joint Convolutional Neural Networks and Context Transfer for Street Scenes Labeling
• Speed/accuracy trade-offs for modern convolutional object detectors
• Fast detection of multiple objects in traffic scenes with a common detection framework
• Multi-View 3D Object Detection Network for Autonomous Driving
• 3D Bounding Box Estimation Using Deep Learning and Geometry

Cited by in corpus (15)

• Towards Large-Scale Small Object Detection: Survey and Benchmarks
• Deep Learning for UAV-based Object Detection and Tracking: A Survey
• DAWN: Vehicle Detection in Adverse Weather Nature Dataset
• Orientation Aware Weapons Detection In Visual Data : A Benchmark Dataset
• Learning a Layout Transfer Network for Context Aware Object Detection
• LF-YOLO: A Lighter and Faster YOLO for Weld Defect Detection of X-ray Image
• Attribute-guided Feature Learning Network for Vehicle Re-identification
• All You Need for Object Detection: From Pixels, Points, and Prompts to Next-Gen Fusion and Multimodal LLMs/VLMs in Autonomous Vehicles
• Semi-MoreGAN: A New Semi-supervised Generative Adversarial Network for Mixture of Rain Removal
• Eliminating cross-camera bias for vehicle re-identification
• Exploiting Global and Local Attentions for Heavy Rain Removal on Single Images
• End-to-end trainable network for degraded license plate detection via vehicle-plate relation mining
• Vehicle Detection in Deep Learning
• KL-Divergence-Based Region Proposal Network for Object Detection
• Pixel-Semantic Revise of Position Learning A One-Stage Object Detector with A Shared Encoder-Decoder