CLUSTSEG: Clustering for Universal Segmentation

James Chenhao Liang, Tianfei Zhou, Dongfang Liu, Wenguan Wang
Proceedings of the 40th International Conference on Machine Learning, PMLR 202:20787-20809, 2023.

Abstract

We present CLUSTSEG, a general, transformer-based framework that tackles different image segmentation tasks ($i.e.,$ superpixel, semantic, instance, and panoptic) through a unified, neural clustering scheme. Regarding queries as cluster centers, CLUSTSEG is innovative in two aspects: 1) cluster centers are initialized in heterogeneous ways so as to pointedly address task-specific demands ($e.g.,$ instance- or category-level distinctiveness), yet without modifying the architecture; and 2) pixel-cluster assignment, formalized in a cross-attention fashion, is alternated with cluster center update, yet without learning additional parameters. These innovations closely link CLUSTSEG to EM clustering and make it a transparent and powerful framework that yields superior results across the above segmentation tasks.

Cite this Paper

BibTeX
@InProceedings{pmlr-v202-liang23h, title = {{CLUSTSEG}: Clustering for Universal Segmentation}, author = {Liang, James Chenhao and Zhou, Tianfei and Liu, Dongfang and Wang, Wenguan}, booktitle = {Proceedings of the 40th International Conference on Machine Learning}, pages = {20787--20809}, year = {2023}, editor = {Krause, Andreas and Brunskill, Emma and Cho, Kyunghyun and Engelhardt, Barbara and Sabato, Sivan and Scarlett, Jonathan}, volume = {202}, series = {Proceedings of Machine Learning Research}, month = {23--29 Jul}, publisher = {PMLR}, pdf = {https://proceedings.mlr.press/v202/liang23h/liang23h.pdf}, url = {https://proceedings.mlr.press/v202/liang23h.html}, abstract = {We present CLUSTSEG, a general, transformer-based framework that tackles different image segmentation tasks ($i.e.,$ superpixel, semantic, instance, and panoptic) through a unified, neural clustering scheme. Regarding queries as cluster centers, CLUSTSEG is innovative in two aspects: 1) cluster centers are initialized in heterogeneous ways so as to pointedly address task-specific demands ($e.g.,$ instance- or category-level distinctiveness), yet without modifying the architecture; and 2) pixel-cluster assignment, formalized in a cross-attention fashion, is alternated with cluster center update, yet without learning additional parameters. These innovations closely link CLUSTSEG to EM clustering and make it a transparent and powerful framework that yields superior results across the above segmentation tasks.} }
Endnote
%0 Conference Paper %T CLUSTSEG: Clustering for Universal Segmentation %A James Chenhao Liang %A Tianfei Zhou %A Dongfang Liu %A Wenguan Wang %B Proceedings of the 40th International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2023 %E Andreas Krause %E Emma Brunskill %E Kyunghyun Cho %E Barbara Engelhardt %E Sivan Sabato %E Jonathan Scarlett %F pmlr-v202-liang23h %I PMLR %P 20787--20809 %U https://proceedings.mlr.press/v202/liang23h.html %V 202 %X We present CLUSTSEG, a general, transformer-based framework that tackles different image segmentation tasks ($i.e.,$ superpixel, semantic, instance, and panoptic) through a unified, neural clustering scheme. Regarding queries as cluster centers, CLUSTSEG is innovative in two aspects: 1) cluster centers are initialized in heterogeneous ways so as to pointedly address task-specific demands ($e.g.,$ instance- or category-level distinctiveness), yet without modifying the architecture; and 2) pixel-cluster assignment, formalized in a cross-attention fashion, is alternated with cluster center update, yet without learning additional parameters. These innovations closely link CLUSTSEG to EM clustering and make it a transparent and powerful framework that yields superior results across the above segmentation tasks.
APA
Liang, J.C., Zhou, T., Liu, D. & Wang, W.. (2023). CLUSTSEG: Clustering for Universal Segmentation. Proceedings of the 40th International Conference on Machine Learning, in Proceedings of Machine Learning Research 202:20787-20809 Available from https://proceedings.mlr.press/v202/liang23h.html.

Related Material