Oriented R-CNN for Object Detection

  title={Oriented R-CNN for Object Detection},
  author={Xingxing Xie and Gong Cheng and Jiabao Wang and Xiwen Yao and Junwei Han},
  journal={2021 IEEE/CVF International Conference on Computer Vision (ICCV)},
Current state-of-the-art two-stage detectors generate oriented proposals through time-consuming schemes. This diminishes the detectors’ speed, thereby becoming the computational bottleneck in advanced oriented object detection systems. This work proposes an effective and simple oriented object detection framework, termed Oriented R-CNN, which is a general two-stage oriented detector with promising accuracy and efficiency. To be specific, in the first stage, we propose an oriented Region… 

Figures and Tables from this paper

Anchor-Free Oriented Proposal Generator for Object Detection

This article proposes a novel anchor-free oriented proposal generator (AOPG) that abandons horizontal box-related operations from the network architecture and first produces coarse oriented boxes by a coarse location module (CLM) in an anchor- free manner and then refines them into high-quality oriented proposals.

Dual-aligned Oriented Detector

Comprehensive and extensive evaluations on three benchmarks, including DIOR-R, DOTA, and HRSC2016, indicate that the two-stage oriented object detection method, termed Dual-aligned Oriented Detector (DODet), could obtain consistent and substantial gains compared with the baseline method.

Dynamic Proposal Generation for Oriented Object Detection in Aerial Images

This work obtains the coarse oriented anchors and utilizes them to align the features and makes further classification and regression to produce final oriented proposals, and designs Maximum Number Estimation (MNE) for predicting an approximate value to remain the proposals adaptively.

AO2-DETR: Arbitrary-Oriented Object Detection Transformer

This paper proposes an Arbitrary-Oriented Object DEtection TRansformer framework, termed AO2-DETR, which comprises three dedicated components and considerably simplifies the overall pipeline and presents a new AOOD paradigm.

Short-Side Excursion for Oriented Object Detection

A novel framework to detect oriented objects, termed short-side excursion detection (SSEDet), inspired by the circle theorem, is proposed, which proposes a transformation method from horizontal rectangles to oriented ones to accurately describe oriented objects.

Point RCNN: An Angle-Free Framework for Rotated Object Detection

It is experimentally verified that re-sampling the images of the rare categories can stabilize the training procedure and further improve the detection performance, and the Point RCNN method achieved new state-of-the-art detection performance on multiple large-scale aerial image datasets, including DOTA-v1.0 and UCAS-AOD.

Object Detection for Aerial Images with Feature Enhancement and Soft Label Assignment

An anchor-free detector with feature enhancement and soft label assignment (FSDet) which adopts a simpler design and achieves competitive performance, and an oriented feature refinement module to align the features with oriented objects to address feature misalignment.

Towards Large-Scale Small Object Detection: Survey and Benchmarks

Two large-scale Small Object Detection dAtasets (SODA), SODA-D and S ODA-A, which focus on the Driving and Aerial scenarios respectively are constructed, and the performance of mainstream methods on SOD a is evaluated.

Task-wise Sampling Convolutions for Arbitrary-Oriented Object Detection in Aerial Images

Extensive experiments on several public datasets covering multiple scenes, multimodal images, and multiple categories of objects demonstrate the effectiveness, scalability and superior performance of the proposed task-wise sampling convolutions (TS-Conv).

LPNet: Retina Inspired Neural Network for Object Detection and Recognition

LPNet improves the detection accuracy by combining retina-like log-polar transformation and stable and sliding modes, and is a plug-and-play architecture for object detection and recognition.



FCOS: Fully Convolutional One-Stage Object Detection

For the first time, a much simpler and flexible detection framework achieving improved detection accuracy is demonstrated, and it is hoped that the proposed FCOS framework can serve as a simple and strong alternative for many other instance-level tasks.

Rich Feature Hierarchies for Accurate Object Detection and Semantic Segmentation

This paper proposes a simple and scalable detection algorithm that improves mean average precision (mAP) by more than 30% relative to the previous best result on VOC 2012 -- achieving a mAP of 53.3%.

Libra R-CNN: Towards Balanced Learning for Object Detection

Libra R-CNN is proposed, a simple but effective framework towards balanced learning for object detection that integrates three novel components: IoU-balanced sampling, balanced feature pyramid, and balanced L1 loss, respectively for reducing the imbalance at sample, feature, and objective level.

Cascade R-CNN: Delving Into High Quality Object Detection

A simple implementation of the Cascade R-CNN is shown to surpass all single-model object detectors on the challenging COCO dataset, and experiments show that it is widely applicable across detector architectures, achieving consistent gains independently of the baseline detector strength.

Learning RoI Transformer for Oriented Object Detection in Aerial Images

The core idea of RoI Transformer is to apply spatial transformations on RoIs and learn the transformation parameters under the supervision of oriented bounding box (OBB) annotations.

Align Deep Features for Oriented Object Detection

A single-shot alignment network (S2A-Net) consisting of two modules: a feature alignment module (FAM) and an oriented detection module (ODM) that can achieve the state-of-the-art performance on two commonly used aerial objects’ data sets while keeping high efficiency.

Fast R-CNN

  • Ross B. Girshick
  • Computer Science, Environmental Science
    2015 IEEE International Conference on Computer Vision (ICCV)
  • 2015
This paper proposes a Fast Region-based Convolutional Network method (Fast R-CNN) for object detection. Fast R-CNN builds on previous work to efficiently classify object proposals using deep

Dynamic Refinement Network for Oriented and Densely Packed Object Detection

A dynamic refinement network that consists of two novel components, i.e., a feature selection module (FSM) and a dynamic refinement head (DRH) that enables neurons to adjust receptive fields in accordance with the shapes and orientations of target objects, which empowers the model to refine the prediction dynamically in an object-aware manner.

Focal Loss for Dense Object Detection

This paper proposes to address the extreme foreground-background class imbalance encountered during training of dense detectors by reshaping the standard cross entropy loss such that it down-weights the loss assigned to well-classified examples, and develops a novel Focal Loss, which focuses training on a sparse set of hard examples and prevents the vast number of easy negatives from overwhelming the detector during training.

Cascade Detector With Feature Fusion For Arbitrary-Oriented Objects In Remote Sensing Images

An end-to-end cascade detector that can effectively detect rotated objects in complex remote sensing images is proposed that achieves superior performance over other state-of-the-art approaches.