Dynamic Refinement Network for Oriented and Densely Packed Object Detection

By Xingjia Pan, Yuqiang Ren, Kekai Sheng, Weiming Dong, Haolei Yuan, Xiao-Wei Guo, Chongyang Ma, Changsheng Xu. Computer Vision and Pattern Recognition, 2020

331 citations recorded by the source corpus.

Object detection has achieved remarkable progress in the past decade. However, the detection of oriented and densely packed objects remains challenging because of following inherent reasons: (1) receptive fields of neurons are all axis-aligned and of the same shape, whereas objects are usually of diverse shapes and align along various directions; (2) detection models are typically trained with generic knowledge and may not generalize well to handle specific objects at test time; (3) the limited dataset hinders the development on this task. To resolve the first two issues, we present a dynamic refinement network that consists of two novel components, i.e., a feature selection module (FSM) and a dynamic refinement head (DRH). Our FSM enables neurons to adjust receptive fields in accordance with the shapes and orientations of target objects, whereas the DRH empowers our model to refine the prediction dynamically in an object-aware manner. To address the limited availability of related benchmarks, we collect an extensive and fully annotated dataset, namely, SKU110K-R, which is relabeled with oriented bounding boxes based on SKU110K. We perform quantitative evaluations on several publicly available benchmarks including DOTA, HRSC2016, SKU110K, and our own SKU110K-R dataset. Experimental results show that our method achieves consistent and substantial gains compared with baseline approaches. Our source code and dataset will be released to encourage follow-up research.

Subfields: Computer Vision and Pattern Recognition, Industrial and Manufacturing Engineering

View Paper: https://www.semanticscholar.org/paper/5e33d0df88356375262bfc98b19ea129bd89e242

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computer Vision and Pattern Recognition

1. Junwei Han
2. Junchi Yan
3. Gong Cheng
4. Xue Yang
5. Guisong Xia
6. Fahad Shahbaz Khan
7. Qi Tian
8. Jianke Zhu

In Industrial and Manufacturing Engineering

1. Xue Yang
2. Qi Tian
3. Jirui Yang
4. Kai Chen
5. Salman Hameed Khan
6. Wentao Wang
7. Gong Cheng
8. Junchi Yan

Citations in the corpus

• Oriented R-CNN for Object Detection (2021, methodology): “As an extended branch of object detection, oriented object detection [7, 37, 26, 12, 5, 28] has received extensive attention driven by its wide applications.”
• Anchor-Free Oriented Proposal Generator for Object Detection (2021, background): “RRPN [6] R2CNN [41] RT [1] GV [2] DRN [21] PIoU [38] DAL [39] S(2)ANet [4] AOPG”
• Learning High-Precision Bounding Box for Rotated Object Detection via Kullback-Leibler Divergence (2021, methodology): “Extending the above horizontal case, existing rotation detection models [1, 12, 13, 14, 15] also use regression loss which simply involves an extra angle parameter θ: tpθ = f(θp − θa), t t θ = f(θt − θa) (3) where f(·) is used to deal with angular periodicity, such as trigonometric functions, modulo”
• ReDet: A Rotation-equivariant Detector for Aerial Object Detection (2021, methodology): “DRN [24] detects oriented objects with dynamic feature selection and refinement.”
• MMRotate: A Rotated Object Detection Benchmark using PyTorch (2022, methodology): “In addition to the three applications mentioned above, rotated object detection is also widely used in 3D objects detection [44] and retail scenes detection [2, 23].”
• Dense Label Encoding for Boundary Discontinuity Free Rotation Detection (2020, unknown): “32 DRN [36] Hourglass104 X 89.”
• R3Det: Refined Single-Stage Detector with Feature Refinement for Rotating Object (2019, methodology): “As for scene text detection, there are many efficient rotation detection methods, such as two-stage methods: R 2 CNN [15], RRPN [ ?”
• Rethinking Rotated Object Detection with Gaussian Wasserstein Distance Loss (2021, methodology): “Arbitrary-oriented objects are ubiquitous for detection across visual datasets, such as aerial images [65, 2, 8, 69], scene text [82, 31, 20, 35, 27], faces [47] and 3D objects [78], retail scenes [5, 41], etc.”
• Align Deep Features for Oriented Object Detection (2020, methodology): “2) Results on HRSC2016: It is noted that DRN [35] and CenterMap-Net [6] are evaluated under PASCAL VOC2012 metrics while other methods are evaluated under PASCAL VOC2007 metrics, and the performance under VOC2012 metrics is better than that under VOC2007 metrics.”
• Transformers in Remote Sensing: A Survey (2022, methodology): “O(2)DETR [47] Detection DOTA [49], SKU110K-R [88], HRSC2016 [81] mAP Extends the standard DETR for oriented detection by introducing an encoder employing depthwise separable convolution.”