# Unbalanced Optimal Transport: A Unified Framework for Object Detection

Source: [Unbalanced Optimal Transport: A Unified Framework for Object Detection](<https://www.sota2.com/research/paper/68fa70021e209131cea48866>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2026-01-04T04:37:04.321Z

Authors: Henri De Plaen, Pierre-Fran\\c{c}ois De Plaen, Johan A. K. Suykens, Marc Proesmans, Tinne Tuytelaars, Luc Van Gool

Published: 2023-07-05T16:21:52.000Z

## Abstract

During training, supervised object detection tries to correctly match the predicted bounding boxes and associated classification scores to the ground truth. This is essential to determine which predictions are to be pushed towards which solutions, or to be discarded. Popular matching strategies include matching to the closest ground truth box (mostly used in combination with anchors), or matching via the Hungarian algorithm (mostly used in anchor-free methods). Each of these strategies comes with its own properties, underlying losses, and heuristics. We show how Unbalanced Optimal Transport unifies these different approaches and opens a whole continuum of methods in between. This allows for a finer selection of the desired properties. Experimentally, we show that training an object detection model with Unbalanced Optimal Transport is able to reach the state-of-the-art both in terms of Average Precision and Average Recall as well as to provide a faster initial convergence. The approach is well suited for GPU implementation, which proves to be an advantage for large-scale models.

## Recorded benchmark results

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

- [Object Detection on Color Boxes synthetic (val)](<https://www.sota2.com/research/sota/object-detection-on-color-boxes-synthetic-val.md>): Object Detection; Color Boxes synthetic (val); AP: 64.

- [Object Detection on COCO 2017 (val)](<https://www.sota2.com/research/sota/object-detection-on-coco-2017-val.md>): Object Detection; COCO 2017 (val); AP: 44.5.

## Original sources and implementation

[Original paper](<https://arxiv.org/abs/2307.02402>)

[PDF](<https://arxiv.org/pdf/2307.02402.pdf>)

[Implementation](<https://hdeplaen.github.io/uotod>)