Learning a Rotation Invariant Detector with Rotatable Bounding Box

By Lei Liu, Zongxu Pan, Bin Lei. 2017

207 citations recorded by the source corpus.

Detection of arbitrarily rotated objects is a challenging task due to the difficulties of locating the multi-angle objects and separating them effectively from the background. The existing methods are not robust to angle varies of the objects because of the use of traditional bounding box, which is a rotation variant structure for locating rotated objects. In this article, a new detection method is proposed which applies the newly defined rotatable bounding box (RBox). The proposed detector (DRBox) can effectively handle the situation where the orientation angles of the objects are arbitrary. The training of DRBox forces the detection networks to learn the correct orientation angle of the objects, so that the rotation invariant property can be achieved. DRBox is tested to detect vehicles, ships and airplanes on satellite images, compared with Faster R-CNN and SSD, which are chosen as the benchmark of the traditional bounding box based methods. The results shows that DRBox performs much better than traditional bounding box based methods do on the given tasks, and is more robust against rotation of input image and target objects. Besides, results show that DRBox correctly outputs the orientation angles of the objects, which is very useful for locating multi-angle objects efficiently. The code and models are available at this https URL

Subfields: Aerospace Engineering, Computer Vision and Pattern Recognition

How researchers see this paper

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

In Aerospace Engineering

1. Xue Yang
2. Junchi Yan
3. Guisong Xia
4. Jian Ding
5. Nan Xue
6. Marco Körner
7. Junwei Han
8. Gong Cheng

In Computer Vision and Pattern Recognition

1. Xue Yang
2. Junchi Yan
3. Guisong Xia
4. Junwei Han
5. Gong Cheng
6. Jian Ding
7. Nan Xue
8. Dacheng Tao

Citations in the corpus

• A Survey of Autonomous Driving: Common Practices and Emerging Technologies (2019, methodology): “Another use of depth image was shown for semantic classification of lidar points [168].”
• ReDet: A Rotation-equivariant Detector for Aerial Object Detection (2021, unknown): “method RC2 [19] RRPN [22] R(2)PN [43] RRD [16] RoI Trans.”
• R3Det: Refined Single-Stage Detector with Feature Refinement for Rotating Object (2019, background): “For in-Figure 1: Performance versus speed on HRSC2016 [26] dataset.”
• RBox-CNN: rotated bounding box based CNN for ship detection in remote sensing image (2018, methodology): “[4] proposed detector using RBox (DRBox) based on SSD and Yang et al.”
• Gliding Vertex on the Horizontal Bounding Box for Multi-Oriented Object Detection (2019, methodology): “, based on horizontal prior boxes, similar with [5], [24], [30], we regress the object center ðx; yÞ, long and short side length ðw0; h0Þ, and the angle u between the long side and X-axis.”
• Learning RoI Transformer for Oriented Object Detection in Aerial Images (2019, methodology): “Therefore, in the design of rotated anchors, both the [30, 24] used a relaxed matching strategy.”
• Object Detection in Optical Remote Sensing Images: A Survey and A New Benchmark (2019, methodology): “…images shows greatly promising performance, remarkable efforts have been made to explore different deep learning based methods (Lin et al., 2017a; Liu et al., 2017a; Liu et al., 2018c; Tang et al., 2017a; Yu et al., 2015; Zou and Shi, 2016), which do not follow the pipeline of region proposal ‐…”
• SCRDet: Towards More Robust Detection for Small, Cluttered and Rotated Objects (2018, unknown): “vehicle [36], ship [41, 42, 28, 43, 27], aircraft [25] etc.”
• Towards Multi-class Object Detection in Unconstrained Remote Sensing Imagery (2018, methodology): “ mining together with concatenated and deconvolutional feature maps. They showed that these methods achieve high accuracies in single-class vehicle detection in aerial images for HBB task. Liu et al. [20] proposed rotated region proposals to predict object orientation using SSD [21] method which imp”
• PIoU Loss: Towards Accurate Oriented Object Detection in Complex Environments (2020, methodology): “owded objects can be better detected and classied. Existing OBB-based approaches are mostly built on anchor-based frameworks by introducing an additional angle dimension optimized by a distance loss [24,18,19,6,41,43] on the parameter tuple (c x;c y;w;h;). While OBB has been primarily used for simpl”
• Learning Modulated Loss for Rotated Object Detection (2019, background): “HRSC2016 (Liu et al. 2017): HRSC2016 is a dataset for ship detection which range of aspect ratio and that of arbitrary orientation are large.”