A Normalized Gaussian Wasserstein Distance for Tiny Object Detection

By Jinwang Wang, Chang Xu, Wen Yang, Lei Yu. arXiv.org, 2021

595 citations recorded by the source corpus.

Detecting tiny objects is a very challenging problem since a tiny object only contains a few pixels in size. We demonstrate that state-of-the-art detectors do not produce satisfactory results on tiny objects due to the lack of appearance information. Our key observation is that Intersection over Union (IoU) based metrics such as IoU itself and its extensions are very sensitive to the location deviation of the tiny objects, and drastically deteriorate the detection performance when used in anchor-based detectors. To alleviate this, we propose a new evaluation metric using Wasserstein distance for tiny object detection. Specifically, we first model the bounding boxes as 2D Gaussian distributions and then propose a new metric dubbed Normalized Wasserstein Distance (NWD) to compute the similarity between them by their corresponding Gaussian distributions. The proposed NWD metric can be easily embedded into the assignment, non-maximum suppression, and loss function of any anchor-based detector to replace the commonly used IoU metric. We evaluate our metric on a new dataset for tiny object detection (AI-TOD) in which the average object size is much smaller than existing object detection datasets. Extensive experiments show that, when equipped with NWD metric, our approach yields performance that is 6.7 AP points higher than a standard fine-tuning baseline, and 6.0 AP points higher than state-of-the-art competitors. Codes are available at: https://github.com/jwwangchn/NWD.

Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition

View Paper: https://www.semanticscholar.org/paper/dcd32548d72640637c447e42ef4d7d468c95f177

How researchers see this paper

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

In Artificial Intelligence

1. Ke Yan
2. Weijun Chen
3. Ziping Yu
4. Hongbo Huang
5. Xiu-Ying Wang
6. Yahui Liu
7. Xiang Yuan
8. Gong Cheng

In Computer Vision and Pattern Recognition

1. Gong Cheng
2. Junwei Han
3. Xiwen Yao
4. Ke Yan
5. Xiang Yuan
6. Qinghua Zeng
7. Pengyu Guo
8. Weijun Chen

Citations in the corpus

• FFCA-YOLO for Small Object Detection in Remote Sensing Images (2024, methodology): “Normalized Wasserstein distance (NWD) [57] loss is added to the loss function of YOLOv5 as a supplement to the box loss.”
• YOLO-FaceV2: A Scale and Occlusion Aware Face Detector (2022, methodology): “Fifth, we borrowed the Normalized Wasserstein Distance metric [30] and introduced it into the regression loss function to balance the shortage of IoU in predicting small faces.”
• Towards Large-Scale Small Object Detection: Survey and Benchmarks (2022, methodology): “Similarly, NWD [135] introduces the Normalized Wasserstein Distance to optimize the location metric for tiny object detectors.”