RON: Reverse Connection with Objectness Prior Networks for Object Detection

Abstract

We present RON, an efficient and effective framework for generic object detection. Our motivation is to smartly associate the best of the region-based (e.g., Faster R-CNN) and region-free (e.g., SSD) methodologies. Under fully convolutional architecture, RON mainly focuses on two fundamental problems: (a) multi-scale object localization and (b) negative sample mining. To address (a), we design the reverse connection, which enables the network to detect objects on multi-levels of CNNs. To deal with (b), we propose the objectness prior to significantly reduce the searching space of objects. We optimize the reverse connection, objectness prior and object detector jointly by a multi-task loss function, thus RON can directly predict final detection results from all locations of various feature maps. Extensive experiments on the challenging PASCAL VOC 2007, PASCAL VOC 2012 and MS COCO benchmarks demonstrate the competitive performance of RON. Specifically, with VGG-16 and low resolution 384X384 input size, the network gets 81.3% mAP on PASCAL VOC 2007, 80.7% mAP on PASCAL VOC 2012 datasets. Its superiority increases when datasets become larger and more difficult, as demonstrated by the results on the MS COCO dataset. With 1.5G GPU memory at test phase, the speed of the network is 15 FPS, 3X faster than the Faster R-CNN counterpart.

Project page will be available at https://github.com/taokong/RON, and formal paper will appear in CVPR 2017

References in corpus (4)

• R-FCN: Object Detection via Region-based Fully Convolutional Networks
• Spatial Pyramid Pooling in Deep Convolutional Networks for Visual Recognition
• Improving Object Detection with Deep Convolutional Networks via Bayesian Optimization and Structured Prediction
• HyperNet: Towards Accurate Region Proposal Generation and Joint Object Detection

Cited by in corpus (19)

• Learning Spatial Fusion for Single-Shot Object Detection
• Deep Learning for UAV-based Object Detection and Tracking: A Survey
• Abdominal multi-organ segmentation with organ-attention networks and statistical fusion
• CornerNet: Detecting Objects as Paired Keypoints
• Review of data analysis in vision inspection of power lines with an in-depth discussion of deep learning technology
• Feature Agglomeration Networks for Single Stage Face Detection
• M2Det: A Single-Shot Object Detector based on Multi-Level Feature Pyramid Network
• Single-Shot Object Detection with Enriched Semantics
• SBNet: Sparse Blocks Network for Fast Inference
• Deep Adaptive Proposal Network for Object Detection in Optical Remote Sensing Images
• Single-Shot Bidirectional Pyramid Networks for High-Quality Object Detection
• ScratchDet: Training Single-Shot Object Detectors from Scratch
• Exploring Multi-Branch and High-Level Semantic Networks for Improving Pedestrian Detection
• Pseudo Mask Augmented Object Detection
• Triply Supervised Decoder Networks for Joint Detection and Segmentation
• Density Map Guided Object Detection in Aerial Images
• A Taught-Obesrve-Ask (TOA) Method for Object Detection with Critical Supervision
• Multiple Attentional Pyramid Networks for Chinese Herbal Recognition
• HR-RCNN: Hierarchical Relational Reasoning for Object Detection