EFDTR: Learnable Elliptical Fourier Descriptor Transformer for Instance Segmentation

Jiawei Cao, Chaochen Gu, Hao Cheng, Xiaofeng Zhang, Kaijie Wu, Changsheng Lu
Proceedings of the 42nd International Conference on Machine Learning, PMLR 267:6543-6553, 2025.

Abstract

Polygon-based object representations efficiently model object boundaries but are limited by high optimization complexity, which hinders their adoption compared to more flexible pixel-based methods. In this paper, we introduce a novel vertex regression loss grounded in Fourier elliptic descriptors, which removes the need for rasterization or heuristic approximations and resolves ambiguities in boundary point assignment through frequency-domain matching. To advance polygon-based instance segmentation, we further propose EFDTR (Elliptical Fourier Descriptor Transformer), an end-to-end learnable framework that leverages the expressiveness of Fourier-based representations. The model achieves precise contour predictions through a two-stage approach: the first stage predicts elliptical Fourier descriptors for global contour modeling, while the second stage refines contours for fine-grained accuracy. Experimental results on the COCO dataset show that EFDTR outperforms existing polygon-based methods, offering a promising alternative to pixel-based approaches. Code is available at https://github.com/chrisclear3/EFDTR.

Cite this Paper

BibTeX
@InProceedings{pmlr-v267-cao25c, title = {{EFDTR}: Learnable Elliptical {F}ourier Descriptor Transformer for Instance Segmentation}, author = {Cao, Jiawei and Gu, Chaochen and Cheng, Hao and Zhang, Xiaofeng and Wu, Kaijie and Lu, Changsheng}, booktitle = {Proceedings of the 42nd International Conference on Machine Learning}, pages = {6543--6553}, year = {2025}, editor = {Singh, Aarti and Fazel, Maryam and Hsu, Daniel and Lacoste-Julien, Simon and Berkenkamp, Felix and Maharaj, Tegan and Wagstaff, Kiri and Zhu, Jerry}, volume = {267}, series = {Proceedings of Machine Learning Research}, month = {13--19 Jul}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v267/main/assets/cao25c/cao25c.pdf}, url = {https://proceedings.mlr.press/v267/cao25c.html}, abstract = {Polygon-based object representations efficiently model object boundaries but are limited by high optimization complexity, which hinders their adoption compared to more flexible pixel-based methods. In this paper, we introduce a novel vertex regression loss grounded in Fourier elliptic descriptors, which removes the need for rasterization or heuristic approximations and resolves ambiguities in boundary point assignment through frequency-domain matching. To advance polygon-based instance segmentation, we further propose EFDTR (Elliptical Fourier Descriptor Transformer), an end-to-end learnable framework that leverages the expressiveness of Fourier-based representations. The model achieves precise contour predictions through a two-stage approach: the first stage predicts elliptical Fourier descriptors for global contour modeling, while the second stage refines contours for fine-grained accuracy. Experimental results on the COCO dataset show that EFDTR outperforms existing polygon-based methods, offering a promising alternative to pixel-based approaches. Code is available at https://github.com/chrisclear3/EFDTR.} }
Endnote
%0 Conference Paper %T EFDTR: Learnable Elliptical Fourier Descriptor Transformer for Instance Segmentation %A Jiawei Cao %A Chaochen Gu %A Hao Cheng %A Xiaofeng Zhang %A Kaijie Wu %A Changsheng Lu %B Proceedings of the 42nd International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2025 %E Aarti Singh %E Maryam Fazel %E Daniel Hsu %E Simon Lacoste-Julien %E Felix Berkenkamp %E Tegan Maharaj %E Kiri Wagstaff %E Jerry Zhu %F pmlr-v267-cao25c %I PMLR %P 6543--6553 %U https://proceedings.mlr.press/v267/cao25c.html %V 267 %X Polygon-based object representations efficiently model object boundaries but are limited by high optimization complexity, which hinders their adoption compared to more flexible pixel-based methods. In this paper, we introduce a novel vertex regression loss grounded in Fourier elliptic descriptors, which removes the need for rasterization or heuristic approximations and resolves ambiguities in boundary point assignment through frequency-domain matching. To advance polygon-based instance segmentation, we further propose EFDTR (Elliptical Fourier Descriptor Transformer), an end-to-end learnable framework that leverages the expressiveness of Fourier-based representations. The model achieves precise contour predictions through a two-stage approach: the first stage predicts elliptical Fourier descriptors for global contour modeling, while the second stage refines contours for fine-grained accuracy. Experimental results on the COCO dataset show that EFDTR outperforms existing polygon-based methods, offering a promising alternative to pixel-based approaches. Code is available at https://github.com/chrisclear3/EFDTR.
APA
Cao, J., Gu, C., Cheng, H., Zhang, X., Wu, K. & Lu, C.. (2025). EFDTR: Learnable Elliptical Fourier Descriptor Transformer for Instance Segmentation. Proceedings of the 42nd International Conference on Machine Learning, in Proceedings of Machine Learning Research 267:6543-6553 Available from https://proceedings.mlr.press/v267/cao25c.html.

Related Material