Light-Head R-CNN: In Defense of Two-Stage Object Detector
Explore this paper's citation graph
Summary
The authors' ResNet-101 based light-head R-CNN outperforms state-of-art object detectors on COCO while keeping time efficiency and significantly outperforming the single-stage, fast detectors like YOLO and SSD on both speed and accuracy.
- Type
- preprint
- Published
- 2017-11-20
- Cited by
- 394
- References
- 40
- Access
- Open access
- OpenAlex
- https://openalex.org/W2769291631
- Semantic Scholar
- https://api.semanticscholar.org/CorpusID:20405715
Keywords
Subnet, Computer science, Detector, Pooling, Computation
References
- Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks
- You Only Look Once: Unified, Real-Time Object Detection
- Very Deep Convolutional Networks for Large-Scale Image Recognition
- Fully convolutional networks for semantic segmentation
- Multiscale Combinatorial Grouping
- Selective Search for Object Recognition
- Going deeper with convolutions
- Rich Feature Hierarchies for Accurate Object Detection and Semantic Segmentation
- Spatial Pyramid Pooling in Deep Convolutional Networks for Visual Recognition
- A wavelet tour of signal processing
- Object Detection Networks on Convolutional Feature Maps
- ImageNet Large Scale Visual Recognition Challenge
- ImageNet classification with deep convolutional neural networks
- Rethinking the Inception Architecture for Computer Vision
- Deep Residual Learning for Image Recognition
- Inception-v4, Inception-ResNet and the Impact of Residual Connections on Learning
- SqueezeNet: AlexNet-level accuracy with 50x fewer parameters and <1MB model size
- Training Region-Based Object Detectors with Online Hard Example Mining
- R-FCN: Object Detection via Region-based Fully Convolutional Networks
- Aggregated Residual Transformations for Deep Neural Networks
Cited by
- Face Detection Using Improved Faster RCNN
- Personalized Attention-Aware Exposure Control Using Reinforcement Learning
- An anchor-free region proposal network for Faster R-CNN-based text detection approaches
- CFENet: An Accurate and Efficient Single-Shot Object Detector for Autonomous Driving
- Non-local RoIs for Instance Segmentation
- Surface defects detection of paper dish based on Mask R-CNN
- Personalized Exposure Control Using Adaptive Metering and Reinforcement Learning
- CornerNet: Detecting Objects as Paired Keypoints
- Hybrid Knowledge Routed Modules for Large-scale Object Detection
- Recent Advances in Object Detection in the Age of Deep Convolutional Neural Networks
- Softer-NMS: Rethinking Bounding Box Regression for Accurate Object Detection
- A Field-Based Representation of Surrounding Vehicle Motion from a Monocular Camera
- Solution for Large-Scale Hierarchical Object Detection Datasets with Incomplete Annotation and Data Imbalance
- Rigid Body Pose Estimation from Line Correspondences
- Hot Anchors: A Heuristic Anchors Sampling Method in RCNN-Based Object Detection
- Comparison Detector: A novel object detection method for small dataset
- Non-local RoI for Cross-Object Perception
- Learning RoI Transformer for Detecting Oriented Objects in Aerial Images
- Decomposed Attention: Self-Attention with Linear Complexities
- AutoFocus: Efficient Multi-Scale Inference
Related papers
- Modeling Complex System Using T-subnet based Hierarchical Petri Nets
- A Class of Petri Nets for Modular and Hierarchical System Modeling
- Type fusion based on multistage neural network
- Subnet Abstract and Transition Refinement in Petri Nets Model
- Study on the Types of EOOPN Subnet Combination
- Taxonomy of Subnets in Petri Net and their Normalization
- Decomposing neural networks into systems of cooperating subnets
- Prioritized Subnet Sampling for Resource-Adaptive Supernet Training
- Subnet Mask Technology and Application