segDeepM: Exploiting Segmentation and Context in Deep Neural Networks for Object Detection

Abstract

In this paper, we propose an approach that exploits object segmentation in order to improve the accuracy of object detection. We frame the problem as inference in a Markov Random Field, in which each detection hypothesis scores object appearance as well as contextual information using Convolutional Neural Networks, and allows the hypothesis to choose and score a segment out of a large pool of accurate object segmentation proposals. This enables the detector to incorporate additional evidence when it is available and thus results in more accurate detections. Our experiments show an improvement of 4.1% in mAP over the R-CNN baseline on PASCAL VOC 2010, and 3.4% over the current state-of-the-art, demonstrating the power of our approach.

References in corpus (4)

• Very Deep Convolutional Networks for Large-Scale Image Recognition
• Efficient Inference in Fully Connected CRFs with Gaussian Edge Potentials
• Going Deeper with Convolutions
• Detect What You Can: Detecting and Representing Objects using Holistic Models and Body Parts

Cited by in corpus (18)

• Deep Learning for Generic Object Detection: A Survey
• Object Detection with Deep Learning: A Review
• 3D Object Proposals using Stereo Imagery for Accurate Object Class Detection
• Visual Semantic Navigation using Scene Priors
• MaskLab: Instance Segmentation by Refining Object Detection with Semantic and Direction Features
• Joint Object and Part Segmentation using Deep Learned Potentials
• Semantic Instance Annotation of Street Scenes by 3D to 2D Label Transfer
• DeePM: A Deep Part-Based Model for Object Detection and Semantic Part Localization
• Full-Resolution Residual Networks for Semantic Segmentation in Street Scenes
• Scene Parsing with Global Context Embedding
• Zoom Better to See Clearer: Human and Object Parsing with Hierarchical Auto-Zoom Net
• A Survey on Deep Domain Adaptation and Tiny Object Detection Challenges, Techniques and Datasets
• Bayesian Relational Memory for Semantic Visual Navigation
• Semantic Video Segmentation: A Review on Recent Approaches
• A Unified Multi-scale Deep Convolutional Neural Network for Fast Object Detection
• Learning to discover and localize visual objects with open vocabulary
• StuffNet: Using 'Stuff' to Improve Object Detection
• TS2C: Tight Box Mining with Surrounding Segmentation Context for Weakly Supervised Object Detection