HR-RCNN: Hierarchical Relational Reasoning for Object Detection

Abstract

Incorporating relational reasoning in neural networks for object recognition remains an open problem. Although many attempts have been made for relational reasoning, they generally only consider a single type of relationship. For example, pixel relations through self-attention (e.g., non-local networks), scale relations through feature fusion (e.g., feature pyramid networks), or object relations through graph convolutions (e.g., reasoning-RCNN). Little attention has been given to more generalized frameworks that can reason across these relationships. In this paper, we propose a hierarchical relational reasoning framework (HR-RCNN) for object detection, which utilizes a novel graph attention module (GAM). This GAM is a concise module that enables reasoning across heterogeneous nodes by operating on the graph edges directly. Leveraging heterogeneous relationships, our HR-RCNN shows great improvement on COCO dataset, for both object detection and instance segmentation.

arXiv commentTo appear at BMVC 2021

References in corpus (23)

• Very Deep Convolutional Networks for Large-Scale Image Recognition
• Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks
• Inductive Representation Learning on Large Graphs
• Efficient Inference in Fully Connected CRFs with Gaussian Edge Potentials
• Conditional Random Fields as Recurrent Neural Networks
• DSSD : Deconvolutional Single Shot Detector
• Learning Combinatorial Optimization Algorithms over Graphs
• High-Resolution Representations for Labeling Pixels and Regions
• Interaction Networks for Learning about Objects, Relations and Physics
• FSSD: Feature Fusion Single Shot Multibox Detector
• Path Aggregation Network for Instance Segmentation
• Beyond Skip Connections: Top-Down Modulation for Object Detection
• Graph networks as learnable physics engines for inference and control
• GCNet: Non-local Networks Meet Squeeze-Excitation Networks and Beyond
• DetectoRS: Detecting Objects with Recursive Feature Pyramid and Switchable Atrous Convolution
• Knowledge Transfer for Out-of-Knowledge-Base Entities: A Graph Neural Network Approach
• Hybrid Knowledge Routed Modules for Large-scale Object Detection
• RON: Reverse Connection with Objectness Prior Networks for Object Detection
• Iterative Visual Reasoning Beyond Convolutions
• Structure Inference Net: Object Detection Using Scene-Level Context and Instance-Level Relationships
• The More You Know: Using Knowledge Graphs for Image Classification
• BorderDet: Border Feature for Dense Object Detection
• Dive Deeper Into Box for Object Detection