PVANET: Deep but Lightweight Neural Networks for Real-time Object Detection

Abstract

This paper presents how we can achieve the state-of-the-art accuracy in multi-category object detection task while minimizing the computational cost by adapting and combining recent technical innovations. Following the common pipeline of "CNN feature extraction + region proposal + RoI classification", we mainly redesign the feature extraction part, since region proposal part is not computationally expensive and classification part can be efficiently compressed with common techniques like truncated SVD. Our design principle is "less channels with more layers" and adoption of some building blocks including concatenated ReLU, Inception, and HyperNet. The designed network is deep and thin and trained with the help of batch normalization, residual connections, and learning rate scheduling based on plateau detection. We obtained solid results on well-known object detection benchmarks: 83.8% mAP (mean average precision) on VOC2007 and 82.5% mAP on VOC2012 (2nd place), while taking only 750ms/image on Intel i7-6700K CPU with a single core and 46ms/image on NVIDIA Titan X GPU. Theoretically, our network requires only 12.3% of the computational cost compared to ResNet-101, the winner on VOC2012.

arXiv commentFull details about "PVANet 9.0" in the VOC2012 leaderboard (https://goo.gl/DuQBku). The test codes are available at https://github.com/sanghoon/pva-faster-rcnn

References in corpus (2)

• R-FCN: Object Detection via Region-based Fully Convolutional Networks
• Fast Algorithms for Convolutional Neural Networks

Cited by in corpus (30)

• TextBoxes++: A Single-Shot Oriented Scene Text Detector
• ShuffleNet: An Extremely Efficient Convolutional Neural Network for Mobile Devices
• Object Detection in 20 Years: A Survey
• SINet: A Scale-insensitive Convolutional Neural Network for Fast Vehicle Detection
• Deep Learning for Generic Object Detection: A Survey
• Speed/accuracy trade-offs for modern convolutional object detectors
• Receptive Field Block Net for Accurate and Fast Object Detection
• EAST: An Efficient and Accurate Scene Text Detector
• Recent Advances in Object Detection in the Age of Deep Convolutional Neural Networks
• Scene Text Detection and Recognition: The Deep Learning Era
• DSOD: Learning Deeply Supervised Object Detectors from Scratch
• PixelLink: Detecting Scene Text via Instance Segmentation
• A Comparison of CNN-based Face and Head Detectors for Real-Time Video Surveillance Applications
• Real-time Scene Text Detection with Differentiable Binarization
• Detecting Multi-Oriented Text with Corner-based Region Proposals
• Scene Text Synthesis for Efficient and Effective Deep Network Training
• Text Detection and Recognition in the Wild: A Review
• Recent Advances in Deep Learning for Object Detection
• Verisimilar Image Synthesis for Accurate Detection and Recognition of Texts in Scenes
• Embedded Real-Time Fall Detection Using Deep Learning For Elderly Care
• Training Bit Fully Convolutional Network for Fast Semantic Segmentation
• Towards High Performance Video Object Detection
• Synthetic Data Generation and Adaption for Object Detection in Smart Vending Machines
• Object Detection from Scratch with Deep Supervision
• Road images augmentation with synthetic traffic signs using neural networks
• GrabAR: Occlusion-aware Grabbing Virtual Objects in AR
• ShuffleDet: Real-Time Vehicle Detection Network in On-board Embedded UAV Imagery
• Efficient Scene Text Detection with Textual Attention Tower
• Continuity-Discrimination Convolutional Neural Network for Visual Object Tracking
• Integrating Deep Learning and Augmented Reality to Enhance Situational Awareness in Firefighting Environments