# Compositor: Bottom-up Clustering and Compositing for Robust Part and Object Segmentation

Source: [Compositor: Bottom-up Clustering and Compositing for Robust Part and Object Segmentation](<https://www.sota2.com/research/paper/68fa715a1e209131cea50c93>)

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

Data updated: 2026-01-04T04:35:00.385Z

Authors: Ju He, Jieneng Chen, Ming-Xian Lin, Qihang Yu, Alan Yuille

Published: 2023-06-12T20:12:02.000Z

## Abstract

In this work, we present a robust approach for joint part and object segmentation. Specifically, we reformulate object and part segmentation as an optimization problem and build a hierarchical feature representation including pixel, part, and object-level embeddings to solve it in a bottom-up clustering manner. Pixels are grouped into several clusters where the part-level embeddings serve as cluster centers. Afterwards, object masks are obtained by compositing the part proposals. This bottom-up interaction is shown to be effective in integrating information from lower semantic levels to higher semantic levels. Based on that, our novel approach Compositor produces part and object segmentation masks simultaneously while improving the mask quality. Compositor achieves state-of-the-art performance on PartImageNet and Pascal-Part by outperforming previous methods by around 0.9% and 1.3% on PartImageNet, 0.4% and 1.7% on Pascal-Part in terms of part and object mIoU and demonstrates better robustness against occlusion by around 4.4% and 7.1% on part and object respectively. Code will be available at https://github.com/TACJu/Compositor.

## 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.

- [Part Segmentation on PartImageNet (val)](<https://www.sota2.com/research/sota/part-segmentation-on-partimagenet-val.md>): Part Segmentation; PartImageNet (val); mIoU: 64.64.

- [Object Segmentation on PartImageNet (val)](<https://www.sota2.com/research/sota/object-segmentation-on-partimagenet-val.md>): Object Segmentation; PartImageNet (val); mIoU: 78.98.

- [Object Segmentation on Pascal-Part (val)](<https://www.sota2.com/research/sota/object-segmentation-on-pascal-part-val.md>): Object Segmentation; Pascal-Part (val); mIoU: 83.1.

- [Part Segmentation on Pascal-Part (val)](<https://www.sota2.com/research/sota/part-segmentation-on-pascal-part-val.md>): Part Segmentation; Pascal-Part (val); mIoU: 55.92.

- [Object Segmentation on Occluded-PartImageNet 20%-40% occlusion v1 (test)](<https://www.sota2.com/research/sota/object-segmentation-on-occluded-partimagenet-20-percent-40-percent-occlusion-v1-test.md>): Object Segmentation; Occluded-PartImageNet 20%-40% occlusion v1 (test); mIoU: 63.79.

- [Part Segmentation on Occluded-PartImageNet 20%-40% occlusion v1 (test)](<https://www.sota2.com/research/sota/part-segmentation-on-occluded-partimagenet-20-percent-40-percent-occlusion-v1-test.md>): Part Segmentation; Occluded-PartImageNet 20%-40% occlusion v1 (test); mIoU: 54.63.

## Original sources and implementation

[Original paper](<https://arxiv.org/abs/2306.07404>)

[PDF](<https://arxiv.org/pdf/2306.07404.pdf>)

[Implementation](<https://github.com/TACJu/Compositor>)