Detecting tiny objects in aerial images: A normalized Wasserstein distance and a new benchmark
By Chang Xu, Jinwang Wang, Wen Yang, Huai Yu, Lei Yu, Guisong Xia. arXiv.org, 2022
369 citations recorded by the source corpus.
Tiny object detection (TOD) in aerial images is challenging since a tiny object only contains a few pixels. State-of-the-art object detectors do not provide satisfactory results on tiny objects due to the lack of supervision from discriminative features. Our key observation is that the Intersection over Union (IoU) metric and its extensions are very sensitive to the location deviation of the tiny objects, which drastically deteriorates the quality of label assignment when used in anchor-based detectors. To tackle this problem, we propose a new evaluation metric dubbed Normalized Wasserstein Distance (NWD) and a new RanKing-based Assigning (RKA) strategy for tiny object detection. The proposed NWD-RKA strategy can be easily embedded into all kinds of anchor-based detectors to replace the standard IoU threshold-based one, significantly improving label assignment and providing sufficient supervision information for network training. Tested on four datasets, NWD-RKA can consistently improve tiny object detection performance by a large margin. Besides, observing prominent noisy labels in the Tiny Object Detection in Aerial Images (AI-TOD) dataset, we are motivated to meticulously relabel it and release AI-TOD-v2 and its corresponding benchmark. In AI-TOD-v2, the missing annotation and location error problems are considerably mitigated, facilitating more reliable training and validation processes. Embedding NWD-RKA into DetectoRS, the detection performance achieves 4.3 AP points improvement over state-of-the-art competitors on AI-TOD-v2. Datasets, codes, and more visualizations are available at: https://chasel-tsui.github.io/AI-TOD-v2/
Subfields: Computer Vision and Pattern Recognition
View Paper: https://www.semanticscholar.org/paper/1e2334fdf1b3d4f1ca099f4c5d2980d7c01a7003
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
Citations in the corpus
- Towards Large-Scale Small Object Detection: Survey and Benchmarks (2022, methodology): “In line with the definition of small or tiny objects in previous works [6], [7], [160], we adopt the absolute area criterion and regard an instance who has an area smaller than 1024 pixels, i.e., S ≤ 1024 , as a Small object.”