PVANet: Lightweight Deep Neural Networks for Real-time Object Detection

Abstract

In object detection, reducing computational cost is as important as improving accuracy for most practical usages. This paper proposes a novel network structure, which is an order of magnitude lighter than other state-of-the-art networks while maintaining the accuracy. Based on the basic principle of more layers with less channels, this new deep neural network minimizes its redundancy by adopting recent innovations including C.ReLU and Inception structure. We also show that this network can be trained efficiently to achieve solid results on well-known object detection benchmarks: 84.9% and 84.2% mAP on VOC2007 and VOC2012 while the required compute is less than 10% of the recent ResNet-101.

Presented at NIPS 2016 Workshop on Efficient Methods for Deep Neural Networks (EMDNN). Continuation of arXiv:1608.08021. The affiliation has been corrected

References in corpus (3)

• Very Deep Convolutional Networks for Large-Scale Image Recognition
• R-FCN: Object Detection via Region-based Fully Convolutional Networks
• Identity Mappings in Deep Residual Networks

Cited by in corpus (10)

• Object Detection with Deep Learning: A Review
• Detecting Multi-Oriented Text with Corner-based Region Proposals
• Mobile Video Object Detection with Temporally-Aware Feature Maps
• An end-to-end TextSpotter with Explicit Alignment and Attention
• Convolutional Character Networks
• PIXOR: Real-time 3D Object Detection from Point Clouds
• BABO: Background Activation Black-Out for Efficient Object Detection
• RUHSNet: 3D Object Detection Using Lidar Data in Real Time
• Vision-based Semantic Mapping and Localization for Autonomous Indoor Parking
• Condensing Two-stage Detection with Automatic Object Key Part Discovery