DetectoRS: Detecting Objects with Recursive Feature Pyramid and Switchable Atrous Convolution

Abstract

Many modern object detectors demonstrate outstanding performances by using the mechanism of looking and thinking twice. In this paper, we explore this mechanism in the backbone design for object detection. At the macro level, we propose Recursive Feature Pyramid, which incorporates extra feedback connections from Feature Pyramid Networks into the bottom-up backbone layers. At the micro level, we propose Switchable Atrous Convolution, which convolves the features with different atrous rates and gathers the results using switch functions. Combining them results in DetectoRS, which significantly improves the performances of object detection. On COCO test-dev, DetectoRS achieves state-of-the-art 55.7% box AP for object detection, 48.5% mask AP for instance segmentation, and 50.0% PQ for panoptic segmentation. The code is made publicly available.

References in corpus (14)

• Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks
• Rethinking Atrous Convolution for Semantic Image Segmentation
• YOLOv3: An Incremental Improvement
• Neural Architecture Search with Reinforcement Learning
• MMDetection: Open MMLab Detection Toolbox and Benchmark
• EfficientDet: Scalable and Efficient Object Detection
• Slimmable Neural Networks
• DeeperLab: Single-Shot Image Parser
• Learning to Fuse Things and Stuff
• Axial-DeepLab: Stand-Alone Axial-Attention for Panoptic Segmentation
• SpineNet: Learning Scale-Permuted Backbone for Recognition and Localization
• SpatialFlow: Bridging All Tasks for Panoptic Segmentation
• Unifying Training and Inference for Panoptic Segmentation
• SOGNet: Scene Overlap Graph Network for Panoptic Segmentation

Cited by in corpus (39)

• Weighted boxes fusion: Ensembling boxes from different object detection models
• A Review on Deep Learning in UAV Remote Sensing
• A Normalized Gaussian Wasserstein Distance for Tiny Object Detection
• Probabilistic two-stage detection
• Scaled-YOLOv4: Scaling Cross Stage Partial Network
• PP-YOLOv2: A Practical Object Detector
• EfficientPose: An efficient, accurate and scalable end-to-end 6D multi object pose estimation approach
• MGNet: Monocular Geometric Scene Understanding for Autonomous Driving
• The SARAS Endoscopic Surgeon Action Detection (ESAD) dataset: Challenges and methods
• Scaling Wide Residual Networks for Panoptic Segmentation
• Implicit Feature Pyramid Network for Object Detection
• A Competitive Method to VIPriors Object Detection Challenge
• 1st Place Solution to ECCV-TAO-2020: Detect and Represent Any Object for Tracking
• GIAOTracker: A comprehensive framework for MCMOT with global information and optimizing strategies in VisDrone 2021
• ViP-DeepLab: Learning Visual Perception with Depth-aware Video Panoptic Segmentation
• Towards large-scale, automated, accurate detection of CCTV camera objects using computer vision. Applications and implications for privacy, safety, and cybersecurity. (Preprint)
• RiWNet: A moving object instance segmentation Network being Robust in adverse Weather conditions
• Centralized Information Interaction for Salient Object Detection
• Locally Enhanced Self-Attention: Combining Self-Attention and Convolution as Local and Context Terms
• Localizing Infinity-shaped fishes: Sketch-guided object localization in the wild
• AQuA: Analytical Quality Assessment for Optimizing Video Analytics Systems
• SRF-GAN: Super-Resolved Feature GAN for Multi-Scale Representation
• Simple multi-dataset detection
• HR-RCNN: Hierarchical Relational Reasoning for Object Detection
• The MIS Check-Dam Dataset for Object Detection and Instance Segmentation Tasks
• VIPriors Object Detection Challenge
• Combinatorial Optimization for Panoptic Segmentation: A Fully Differentiable Approach
• Transformer Assisted Convolutional Network for Cell Instance Segmentation
• DeepDoseNet: A Deep Learning model for 3D Dose Prediction in Radiation Therapy
• Perception-and-Regulation Network for Salient Object Detection
• Global Context Aware RCNN for Object Detection
• DeepRepair: Style-Guided Repairing for DNNs in the Real-world Operational Environment
• CoDeNet: Efficient Deployment of Input-Adaptive Object Detection on Embedded FPGAs
• Panoptic Segmentation: A Review
• 1st Place Solutions for UG2+ Challenge 2021 -- (Semi-)supervised Face detection in the low light condition
• Zero Cost Improvements for General Object Detection Network
• ACFNet: Adaptively-Cooperative Fusion Network for RGB-D Salient Object Detection
• USB: Universal-Scale Object Detection Benchmark
• 3rd Place Scheme on Instance Segmentation Track of ICCV 2021 VIPriors Challenges