Tree-Structured Reinforcement Learning for Sequential Object Localization

Abstract

Existing object proposal algorithms usually search for possible object regions over multiple locations and scales separately, which ignore the interdependency among different objects and deviate from the human perception procedure. To incorporate global interdependency between objects into object localization, we propose an effective Tree-structured Reinforcement Learning (Tree-RL) approach to sequentially search for objects by fully exploiting both the current observation and historical search paths. The Tree-RL approach learns multiple searching policies through maximizing the long-term reward that reflects localization accuracies over all the objects. Starting with taking the entire image as a proposal, the Tree-RL approach allows the agent to sequentially discover multiple objects via a tree-structured traversing scheme. Allowing multiple near-optimal policies, Tree-RL offers more diversity in search paths and is able to find multiple objects with a single feed-forward pass. Therefore, Tree-RL can better cover different objects with various scales which is quite appealing in the context of object proposal. Experiments on PASCAL VOC 2007 and 2012 validate the effectiveness of the Tree-RL, which can achieve comparable recalls with current object proposal algorithms via much fewer candidate windows.

Advances in Neural Information Processing Systems 2016

References in corpus (7)

• Very Deep Convolutional Networks for Large-Scale Image Recognition
• Playing Atari with Deep Reinforcement Learning
• Recurrent Models of Visual Attention
• What makes for effective detection proposals?
• Multiple Object Recognition with Visual Attention
• Adaptive Object Detection Using Adjacency and Zoom Prediction
• Multiple Instance Reinforcement Learning for Efficient Weakly-Supervised Detection in Images

Cited by in corpus (18)

• Deep Reinforcement Learning: An Overview
• Too Far to See? Not Really! --- Pedestrian Detection with Scale-aware Localization Policy
• Deep Self-Taught Learning for Weakly Supervised Object Localization
• Deep Reinforcement Learning in Computer Vision: A Comprehensive Survey
• Counterfactual Critic Multi-Agent Training for Scene Graph Generation
• PixelRL: Fully Convolutional Network with Reinforcement Learning for Image Processing
• End-to-end Active Object Tracking and Its Real-world Deployment via Reinforcement Learning
• A Self-Adaptive Proposal Model for Temporal Action Detection based on Reinforcement Learning
• CIRL: Controllable Imitative Reinforcement Learning for Vision-based Self-driving
• Reinforcement Learning with Efficient Active Feature Acquisition
• Learning a Reinforced Agent for Flexible Exposure Bracketing Selection
• Quality-Aware Multimodal Saliency Detection via Deep Reinforcement Learning
• RLTIR: Activity-based Interactive Person Identification based on Reinforcement Learning Tree
• Learning to Navigate for Fine-grained Classification
• Cost-Aware Fine-Grained Recognition for IoTs Based on Sequential Fixations
• Sequential Voting with Relational Box Fields for Active Object Detection
• Adaptive Remote Sensing Image Attribute Learning for Active Object Detection
• Weakly Supervised Object Localization and Detection: A Survey