A curated list of awesome resources for dichotomous image segmentation (DIS).
67
71 commits
updated Jun 16, 2026
README
π― Awesome Dichotomous Image Segmentation
π₯ A curated list of awesome resources for dichotomous image segmentation (DIS)
π¨π³ δΈζη | English
π’ Latest Updates
| Date | News |
|---|---|
| π₯ 2026/05/25 | CVPR2026: FlowDIS - Language-Guided DIS with Flow Matching |
| π₯ 2026/03/05 | CVPR2026: PDFNet - High-Precision DIS via Depth Integrity-Prior |
| π₯ 2026/03/05 | ICLR: S3OD - Generalizable SOD with Synthetic Data |
| π₯ 2026/02/26 | Sensors: PMG-SAM - Boosting SAM with Pre-Mask Guidance |
| π₯ 2025/11/19 | arXiv: S3OD - Towards Generalizable Salient Object Detection |
| π₯ 2025/10/24 | arXiv: M2N2V2 - Multi-Modal Unsupervised Interactive Segmentation |
π View More Updates
| Date | News |
|---|---|
| π₯ 2025/09/15 | arXiv: SAM2-UNeXT - Adapting Foundation Models to Downstream Segmentation |
| π₯ 2025/08/05 | ICCV: LawDIS - Language-Window-based Controllable DIS |
| π₯ 2025/07/15 | PR: DC-Net - Divide-and-conquer for salient object detection |
| π₯ 2025/05/25 | arXiv: MGD-SAM2 - Multi-view Guided Detail-enhanced SAM 2 |
| π₯ 2025/04/19 | ICLR: OrderIS - Order-aware Interactive Segmentation |
| π₯ 2025/03/24 | ICME: DIS-SAM - Promoting SAM towards Highly Accurate DIS |
| π₯ 2025/03/15 | arXiv: High-Precision DIS via Depth Integrity-Prior |
| π₯ 2025/03/10 | PR: S2DiNet - Lightweight and Fast High-Resolution DIS |
| π₯ 2025/02/20 | ESWA: EG-SAM - Edge-Guided SAM for Complex Object Segmentation |
| π₯ 2025/01/17 | arXiv: BEN - Confidence-Guided Matting for DIS |
| π₯ 2024/12/15 | Diffusion Models paper updated |
| π₯ 2024/11/16 | arXiv: High-Precision DIS via Probing Diffusion Capacity |
| π₯ 2024/10/12 | NeurIPS: MaskFactory - Synthetic Data Generation For DIS |
| π₯ 2024/09/11 | CVIU: DCENet - Dual cross-enhancement network |
| π₯ 2024/08/27 | π Created the list |
π οΈ ComfyUI Integration
Want to try these models in ComfyUI? Check out ComfyUI-RemoveBackground_SET by Salvador E. Tropea!
Big thanks to him for making this possible! β€οΈ
π¨ What is DIS?
Dichotomous Image Segmentation (DIS) aims to segment objects from natural images with high precision. Unlike traditional segmentation, DIS focuses on extracting fine-grained details and accurate boundaries.
π Table of Contents
π€ Non-Promptable Methods
Preprint
| Year | Pub. | ποΈ Network | Title | Author | Links |
|---|---|---|---|---|---|
| 2025 | arXiv | BEN | Using Confidence-Guided Matting for DIS | Maxwell Meyer, Jack Spruyt | Paper / Code  |
| Leverages confidence-guided matting techniques to improve segmentation quality in ambiguous boundary regions. |
2026
| Year | Pub. | ποΈ Network | Title | Author | Links |
|---|---|---|---|---|---|
| 2026 | CVPR | FlowDIS | Language-Guided Dichotomous Image Segmentation with Flow Matching | Andranik Sargsyan, Shant Navasardyan | Paper / Project / Code  |
| Builds on flow matching to transport the image distribution to the corresponding mask distribution, with optional text-prompt guidance via the Position-Aware Instance Pairing (PAIP) training strategy. | |||||
| 2026 | CVPR | PDFNet | High-Precision DIS via Depth Integrity-Prior and Fine-Grained Patch Strategy | Xianjie Liu, Keren Fu, Qijun Zhao | Paper / Code  |
| Introduces depth integrity prior and fine-grained patch strategy to achieve high-precision segmentation for complex objects. | |||||
| 2026 | ICLR | S3OD | Towards Generalizable Salient Object Detection with Synthetic Data | Orest Kupyn, Hirokatsu Kataoka, Christian Rupprecht | Paper / Project |
| Addresses the generalization problem in salient object detection by leveraging high-quality synthetic training data. |
2025
| Year | Pub. | ποΈ Network | Title | Author | Links |
|---|---|---|---|---|---|
| 2025 | ICLR | DiffDIS | High-Precision DIS via Probing Diffusion Capacity | Qian Yu, Peng-Tao Jiang, Hao Zhang, Jinwei Chen, Bo Li, Lihe Zhang, Huchuan Lu | Paper / Code  |
| Explores the inherent capacity of pre-trained diffusion models for high-precision dichotomous image segmentation. | |||||
| 2025 | ICLR | GenPercept | What Matters When Repurposing Diffusion Models for General Dense Perception Tasks? | Guangkai Xu, Yongtao Ge, Mingyu Liu, Chengxiang Fan, Kangyang Xie, Zhiyue Zhao, Hao Chen, Chunhua Shen | Paper / Code  |
| Systematically investigates key factors when adapting diffusion models for dense prediction tasks including DIS. | |||||
| 2025 | PR | S2DiNet | Towards Lightweight and Fast High-Resolution DIS | Shuhan Chen, Haonan Tang, Yuan Huang, Lifeng Zhang, Xuelong Hu | Paper / Code  |
| Proposes a lightweight and efficient network architecture for high-resolution DIS with fast inference speed. |
2024
| Year | Pub. | ποΈ Network | Title | Author | Links |
|---|---|---|---|---|---|
| 2024 | NeurIPS | MaskFactory | Towards High-quality Synthetic Data Generation For DIS | Haotian Qian, YD Chen, Shengtao Lou, Fahad Shahbaz Khan, Xiaogang Jin, Deng-Ping Fan | Paper / Code  |
| Generates high-quality synthetic training data for DIS by combining image compositing and mask refinement techniques. | |||||
| 2024 | CVPR | MVANet | Multi-view Aggregation Network for DIS | Qian Yu, Xiaoqi Zhao, Youwei Pang, Lihe Zhang, Huchuan Lu | Paper / Code  |
| Proposes multi-view aggregation strategy to capture both global context and local details for accurate segmentation. | |||||
| 2024 | CAAI AIR | BiRefNet | Bilateral Reference for High-Resolution DIS | Peng Zheng, Dehong Gao, Deng-Ping Fan, Li Liu, Jorma Laaksonen, Wanli Ouyang, Nicu Sebe | Paper / Code  |
| Introduces bilateral reference mechanism to handle high-resolution images efficiently while preserving fine details. | |||||
| 2024 | TNNLS | FSANet | High-Precision DIS With Frequency and Scale Awareness | Qiuping Jiang, Jinguang Cheng, Zongwei Wu, Runmin Cong, Radu Timofte | Paper / Code  |
| Incorporates frequency domain analysis and multi-scale features to achieve high-precision boundary segmentation. | |||||
| 2024 | CVIU | DCENet | Dual Cross-enhancement Network for Highly Accurate DIS | Hongbo Bi, Yuyu Tong, Pan Zhang, Jiayuan Zhang, Cong Zhang | Paper / Code  |
| Designs dual cross-enhancement modules to mutually reinforce feature representations for accurate segmentation. | |||||
| 2024 | TVC | BA-DIS | Boundary-aware Dichotomous Image Segmentation | Haonan Tang, Shuhan Chen, Yang Liu, Shiyu Wang, Zeyu Chen, Xuelong Hu | Paper / Code  |
| Focuses on boundary-aware learning to improve edge accuracy in dichotomous image segmentation. | |||||
| 2024 | PR | DC-Net | Divide-and-conquer for Salient Object Detection | Jiayi Zhu, Xuebin Qin, Abdulmotaleb Elsaddik | Paper / Code  |
| Adopts divide-and-conquer strategy to handle objects at different scales for robust salient object detection. |
2023
| Year | Pub. | ποΈ Network | Title | Author | Links |
|---|---|---|---|---|---|
| 2023 | ACM MM | UDUN | Unite-Divide-Unite: Joint Boosting Trunk and Structure for High-accuracy DIS | Jialun Pei, Zhangjun Zhou, Yueming Jin, He Tang, Pheng-Ann Heng | Paper / Code  |
| Proposes unite-divide-unite paradigm to jointly optimize trunk features and structural details for high accuracy. | |||||
| 2023 | IJCAI | FP-DIS | Dichotomous Image Segmentation with Frequency Priors | Yan Zhou, Bo Dong, Yuanfeng Wu, Wentao Zhu, Geng Chen, Yanning Zhang | Paper / Code  |
| Leverages frequency domain priors to enhance boundary perception and segmentation accuracy. |
2022
| Year | Pub. | ποΈ Network | Title | Author | Links |
|---|---|---|---|---|---|
| 2022 | ECCV | IS-Net | Highly Accurate Dichotomous Image Segmentation | Xuebin Qin, Hang Dai, Xiaobin Hu, Deng-Ping Fan, Ling Shao, Luc Van Gool | Paper / Code  |
| π± The pioneering work that defines the DIS task and proposes a baseline with a large-scale dataset. |
π Promptable Methods
π€ Methods based on SAM or other foundation models with interactive prompts
Preprint
| Year | Pub. | ποΈ Network | Title | Author | Links |
|---|---|---|---|---|---|
| 2025 | arXiv | M2N2V2 | Multi-Modal Unsupervised and Training-free Interactive Segmentation | Markus Karmann, Peng-Tao Jiang, Bo Li, Onay Urfalioglu | Paper |
| Proposes a training-free approach for interactive segmentation using multi-modal information without additional training. | |||||
| 2025 | arXiv | SAM2-UNeXT | An Improved High-Resolution Baseline for Adapting Foundation Models | Xinyu Xiong, Zihuang Wu, Lei Zhang, Lei Lu, Ming Li, Guanbin Li | Paper / Code  |
| Develops an improved high-resolution baseline for adapting SAM2 to downstream segmentation tasks. | |||||
| 2026 | IEEE TCSVT | MGD-SAM2 | Multi-view Guided Detail-enhanced SAM 2 for High-Resolution Segmentation | Haoran Shen, Peixian Zhuang, Jiahao Kou, Yuxin Zeng, Haoying Xu, Jiangyun Li | Paper / Code  |
| Enhances SAM2 with multi-view guidance and detail enhancement for high-resolution class-agnostic segmentation. | |||||
| 2025 | ICME | SU-SAM | A Simple Unified Framework for Adapting SAM in Underperformed Scenes | Yiran Song, Qianyu Zhou, Xuequan Lu, Zhiwen Shao, Lizhuang Ma | Paper / Code  |
| Proposes a simple unified framework to adapt SAM for challenging scenes where the original model underperforms. |
2026
| Year | Pub. | ποΈ Network | Title | Author | Links |
|---|---|---|---|---|---|
| 2026 | CVPR | FlowDIS | Language-Guided Dichotomous Image Segmentation with Flow Matching | Andranik Sargsyan, Shant Navasardyan | Paper / Project / Code |
| Builds on flow matching to transport the image distribution to the corresponding mask distribution, with optional text-prompt guidance via the Position-Aware Instance Pairing (PAIP) training strategy. | |||||
| 2026 | Sensors | PMG-SAM | Boosting Auto-Segmentation of SAM with Pre-Mask Guidance | Jixue Gao, Xiaoyan Jiang, Anjie Wang, Yongbin Gao, Zhijun Fang, Michael S. Lew | Paper |
| Introduces pre-mask guidance mechanism to boost SAM's auto-segmentation capability for better initial predictions. |
2025
| Year | Pub. | ποΈ Network | Title | Author | Links |
|---|---|---|---|---|---|
| 2025 | ICCV | LawDIS | Language-Window-based Controllable DIS | Xinyu Yan, Meijun Sun, Ge-Peng Ji, Fahad Shahbaz Khan, Salman Khan, Deng-Ping Fan | Paper / Code  |
| Enables controllable DIS through language and window-based interaction for flexible user control. | |||||
| 2025 | ICLR | OrderIS | Order-aware Interactive Segmentation | Bin Wang, Anwesa Choudhuri, Meng Zheng, Zhongpai Gao, Benjamin Planche, Andong Deng, Qin Liu, Terrence Chen, Ulas Bagci, Ziyan Wu | Paper |
| Incorporates order-aware learning into interactive segmentation to model the sequential nature of user interactions. | |||||
| 2025 | ICME | DIS-SAM | Promoting SAM towards Highly Accurate DIS | Xianjie Liu, Keren Fu, Yao Jiang, Qijun Zhao | Paper / Code  |
| Adapts SAM specifically for high-precision DIS with enhanced boundary awareness and detail preservation. | |||||
| 2025 | ESWA | EG-SAM | An Edge-Guided SAM for Effective Complex Object Segmentation | Longyi Chen, Xiandong Wang, Fengqin Yao, Mingchen Song, Jiaheng Zhang, Shengke Wang | Paper / Code  |
| Integrates edge guidance into SAM for more effective segmentation of complex objects with intricate boundaries. |
2024
| Year | Pub. | ποΈ Network | Title | Author | Links |
|---|---|---|---|---|---|
| 2024 | ACM MM | PI-SAM | Segment Anything with Precise Interaction | Mengzhen Liu, Mengyu Wang, Henghui Ding, Yilong Xu, Yao Zhao, Yunchao Wei | Paper |
| Improves SAM's interaction mechanism for more precise user control and better segmentation accuracy. | |||||
| 2024 | CVPR | SegNext | Rethinking Interactive Image Segmentation with Low Latency High Quality | Qin Liu, Jaemin Cho, Mohit Bansal, Marc Niethammer | Paper / Code  |
| Rethinks the trade-off between latency and quality in interactive segmentation for practical applications. | |||||
| 2024 | ECCV | CAT-SAM | Conditional Tuning for Few-Shot Adaptation of SAM | Aoran Xiao, Weihao Xuan, Heli Qi, Yun Xing, Ruijie Ren, Xiaoqin Zhang, Ling Shao, Shijian Lu | Paper / Code  |
| Proposes conditional tuning approach for few-shot adaptation of SAM to new domains with limited samples. | |||||
| 2024 | ICCV | BA-SAM | Scalable Bias-Mode Attention Mask for SAM | Yiran Song, Qianyu Zhou, Xiangtai Li, Deng-Ping Fan, Xuequan Lu, Lizhuang Ma | Paper |
| Introduces bias-mode attention mask to improve SAM's segmentation quality with scalable design. | |||||
| 2024 | ECCV | OVSAM | Mamba or RWKV: Exploring High-Quality and High-Efficiency SAM | Haobo Yuan, Xiangtai Li, Lu Qi, Tao Zhang, Ming-Hsuan Yang, Shuicheng Yan, Chen Change Loy | Paper / Code  |
| Explores Mamba and RWKV architectures as alternatives to transformers for efficient high-quality segmentation. | |||||
| 2024 | ICME | PA-SAM | Prompt Adapter SAM for High-Quality Image Segmentation | Zhaozhi Xie, Bochen Guan, Weihao Jiang, Muyang Yi, Yue Ding, Hongtao Lu, Lei Zhang | Paper / Code  |
| Designs a prompt adapter to enhance SAM's capability for high-quality image segmentation tasks. | |||||
| 2024 | PRICAI | BSRNet | Prior Mask-Guided Highly Accurate DIS | Shanfeng Zhou, Bo Yuan, Keren Fu, Hailun Zhang, Qijun Zhao | Paper / Code  |
| Utilizes prior mask guidance to achieve highly accurate dichotomous image segmentation. |
2023
| Year | Pub. | ποΈ Network | Title | Author | Links |
|---|---|---|---|---|---|
| 2023 | NeurIPS | SAM-HQ | Segment Anything in High Quality | Lei Ke, Mingqiao Ye, Martin Danelljan, Yifan Liu, Yu-Wing Tai, Chi-Keung Tang, Fisher Yu | Paper / Code  |
| π Enhances SAM with high-quality output for finer boundaries and more accurate segmentation masks. | |||||
| 2023 | NeurIPS | SegRefiner | Towards Model-Agnostic Segmentation Refinement with Discrete Diffusion | Mengyu Wang, Henghui Ding, Jun Hao Liew, Jiajun Liu, Yao Zhao, Yunchao Wei | Paper / Code  |
| Proposes a model-agnostic refinement approach using discrete diffusion to improve segmentation quality. |
β Star History
If you find this repository useful, please consider giving it a star β!
π License
This project is licensed under the MIT License - see the LICENSE file for details.
Made with β€οΈ for the DIS community
Contributors
Tennine2077
70 commits
AndranikSargsyan
1 commits
Similar repos
A curated list of awesome resources for dichotomous image segmentation (DIS).
67
71 commits
updated Jun 16, 2026
README
π― Awesome Dichotomous Image Segmentation
π₯ A curated list of awesome resources for dichotomous image segmentation (DIS)
π¨π³ δΈζη | English
π’ Latest Updates
| Date | News |
|---|---|
| π₯ 2026/05/25 | CVPR2026: FlowDIS - Language-Guided DIS with Flow Matching |
| π₯ 2026/03/05 | CVPR2026: PDFNet - High-Precision DIS via Depth Integrity-Prior |
| π₯ 2026/03/05 | ICLR: S3OD - Generalizable SOD with Synthetic Data |
| π₯ 2026/02/26 | Sensors: PMG-SAM - Boosting SAM with Pre-Mask Guidance |
| π₯ 2025/11/19 | arXiv: S3OD - Towards Generalizable Salient Object Detection |
| π₯ 2025/10/24 | arXiv: M2N2V2 - Multi-Modal Unsupervised Interactive Segmentation |
π View More Updates
| Date | News |
|---|---|
| π₯ 2025/09/15 | arXiv: SAM2-UNeXT - Adapting Foundation Models to Downstream Segmentation |
| π₯ 2025/08/05 | ICCV: LawDIS - Language-Window-based Controllable DIS |
| π₯ 2025/07/15 | PR: DC-Net - Divide-and-conquer for salient object detection |
| π₯ 2025/05/25 | arXiv: MGD-SAM2 - Multi-view Guided Detail-enhanced SAM 2 |
| π₯ 2025/04/19 | ICLR: OrderIS - Order-aware Interactive Segmentation |
| π₯ 2025/03/24 | ICME: DIS-SAM - Promoting SAM towards Highly Accurate DIS |
| π₯ 2025/03/15 | arXiv: High-Precision DIS via Depth Integrity-Prior |
| π₯ 2025/03/10 | PR: S2DiNet - Lightweight and Fast High-Resolution DIS |
| π₯ 2025/02/20 | ESWA: EG-SAM - Edge-Guided SAM for Complex Object Segmentation |
| π₯ 2025/01/17 | arXiv: BEN - Confidence-Guided Matting for DIS |
| π₯ 2024/12/15 | Diffusion Models paper updated |
| π₯ 2024/11/16 | arXiv: High-Precision DIS via Probing Diffusion Capacity |
| π₯ 2024/10/12 | NeurIPS: MaskFactory - Synthetic Data Generation For DIS |
| π₯ 2024/09/11 | CVIU: DCENet - Dual cross-enhancement network |
| π₯ 2024/08/27 | π Created the list |
π οΈ ComfyUI Integration
Want to try these models in ComfyUI? Check out ComfyUI-RemoveBackground_SET by Salvador E. Tropea!
Big thanks to him for making this possible! β€οΈ
π¨ What is DIS?
Dichotomous Image Segmentation (DIS) aims to segment objects from natural images with high precision. Unlike traditional segmentation, DIS focuses on extracting fine-grained details and accurate boundaries.
π Table of Contents
π€ Non-Promptable Methods
Preprint
| Year | Pub. | ποΈ Network | Title | Author | Links |
|---|---|---|---|---|---|
| 2025 | arXiv | BEN | Using Confidence-Guided Matting for DIS | Maxwell Meyer, Jack Spruyt | Paper / Code |
| Leverages confidence-guided matting techniques to improve segmentation quality in ambiguous boundary regions. |
2026
| Year | Pub. | ποΈ Network | Title | Author | Links |
|---|---|---|---|---|---|
| 2026 | CVPR | FlowDIS | Language-Guided Dichotomous Image Segmentation with Flow Matching | Andranik Sargsyan, Shant Navasardyan | Paper / Project / Code |
| Builds on flow matching to transport the image distribution to the corresponding mask distribution, with optional text-prompt guidance via the Position-Aware Instance Pairing (PAIP) training strategy. | |||||
| 2026 | CVPR | PDFNet | High-Precision DIS via Depth Integrity-Prior and Fine-Grained Patch Strategy | Xianjie Liu, Keren Fu, Qijun Zhao | Paper / Code |
| Introduces depth integrity prior and fine-grained patch strategy to achieve high-precision segmentation for complex objects. | |||||
| 2026 | ICLR | S3OD | Towards Generalizable Salient Object Detection with Synthetic Data | Orest Kupyn, Hirokatsu Kataoka, Christian Rupprecht | Paper / Project |
| Addresses the generalization problem in salient object detection by leveraging high-quality synthetic training data. |
2025
| Year | Pub. | ποΈ Network | Title | Author | Links |
|---|---|---|---|---|---|
| 2025 | ICLR | DiffDIS | High-Precision DIS via Probing Diffusion Capacity | Qian Yu, Peng-Tao Jiang, Hao Zhang, Jinwei Chen, Bo Li, Lihe Zhang, Huchuan Lu | Paper / Code |
| Explores the inherent capacity of pre-trained diffusion models for high-precision dichotomous image segmentation. | |||||
| 2025 | ICLR | GenPercept | What Matters When Repurposing Diffusion Models for General Dense Perception Tasks? | Guangkai Xu, Yongtao Ge, Mingyu Liu, Chengxiang Fan, Kangyang Xie, Zhiyue Zhao, Hao Chen, Chunhua Shen | Paper / Code |
| Systematically investigates key factors when adapting diffusion models for dense prediction tasks including DIS. | |||||
| 2025 | PR | S2DiNet | Towards Lightweight and Fast High-Resolution DIS | Shuhan Chen, Haonan Tang, Yuan Huang, Lifeng Zhang, Xuelong Hu | Paper / Code |
| Proposes a lightweight and efficient network architecture for high-resolution DIS with fast inference speed. |
2024
| Year | Pub. | ποΈ Network | Title | Author | Links |
|---|---|---|---|---|---|
| 2024 | NeurIPS | MaskFactory | Towards High-quality Synthetic Data Generation For DIS | Haotian Qian, YD Chen, Shengtao Lou, Fahad Shahbaz Khan, Xiaogang Jin, Deng-Ping Fan | Paper / Code |
| Generates high-quality synthetic training data for DIS by combining image compositing and mask refinement techniques. | |||||
| 2024 | CVPR | MVANet | Multi-view Aggregation Network for DIS | Qian Yu, Xiaoqi Zhao, Youwei Pang, Lihe Zhang, Huchuan Lu | Paper / Code |
| Proposes multi-view aggregation strategy to capture both global context and local details for accurate segmentation. | |||||
| 2024 | CAAI AIR | BiRefNet | Bilateral Reference for High-Resolution DIS | Peng Zheng, Dehong Gao, Deng-Ping Fan, Li Liu, Jorma Laaksonen, Wanli Ouyang, Nicu Sebe | Paper / Code |
| Introduces bilateral reference mechanism to handle high-resolution images efficiently while preserving fine details. | |||||
| 2024 | TNNLS | FSANet | High-Precision DIS With Frequency and Scale Awareness | Qiuping Jiang, Jinguang Cheng, Zongwei Wu, Runmin Cong, Radu Timofte | Paper / Code |
| Incorporates frequency domain analysis and multi-scale features to achieve high-precision boundary segmentation. | |||||
| 2024 | CVIU | DCENet | Dual Cross-enhancement Network for Highly Accurate DIS | Hongbo Bi, Yuyu Tong, Pan Zhang, Jiayuan Zhang, Cong Zhang | Paper / Code |
| Designs dual cross-enhancement modules to mutually reinforce feature representations for accurate segmentation. | |||||
| 2024 | TVC | BA-DIS | Boundary-aware Dichotomous Image Segmentation | Haonan Tang, Shuhan Chen, Yang Liu, Shiyu Wang, Zeyu Chen, Xuelong Hu | Paper / Code |
| Focuses on boundary-aware learning to improve edge accuracy in dichotomous image segmentation. | |||||
| 2024 | PR | DC-Net | Divide-and-conquer for Salient Object Detection | Jiayi Zhu, Xuebin Qin, Abdulmotaleb Elsaddik | Paper / Code |
| Adopts divide-and-conquer strategy to handle objects at different scales for robust salient object detection. |
2023
| Year | Pub. | ποΈ Network | Title | Author | Links |
|---|---|---|---|---|---|
| 2023 | ACM MM | UDUN | Unite-Divide-Unite: Joint Boosting Trunk and Structure for High-accuracy DIS | Jialun Pei, Zhangjun Zhou, Yueming Jin, He Tang, Pheng-Ann Heng | Paper / Code |
| Proposes unite-divide-unite paradigm to jointly optimize trunk features and structural details for high accuracy. | |||||
| 2023 | IJCAI | FP-DIS | Dichotomous Image Segmentation with Frequency Priors | Yan Zhou, Bo Dong, Yuanfeng Wu, Wentao Zhu, Geng Chen, Yanning Zhang | Paper / Code |
| Leverages frequency domain priors to enhance boundary perception and segmentation accuracy. |
2022
| Year | Pub. | ποΈ Network | Title | Author | Links |
|---|---|---|---|---|---|
| 2022 | ECCV | IS-Net | Highly Accurate Dichotomous Image Segmentation | Xuebin Qin, Hang Dai, Xiaobin Hu, Deng-Ping Fan, Ling Shao, Luc Van Gool | Paper / Code |
| π± The pioneering work that defines the DIS task and proposes a baseline with a large-scale dataset. |
π Promptable Methods
π€ Methods based on SAM or other foundation models with interactive prompts
Preprint
| Year | Pub. | ποΈ Network | Title | Author | Links |
|---|---|---|---|---|---|
| 2025 | arXiv | M2N2V2 | Multi-Modal Unsupervised and Training-free Interactive Segmentation | Markus Karmann, Peng-Tao Jiang, Bo Li, Onay Urfalioglu | Paper |
| Proposes a training-free approach for interactive segmentation using multi-modal information without additional training. | |||||
| 2025 | arXiv | SAM2-UNeXT | An Improved High-Resolution Baseline for Adapting Foundation Models | Xinyu Xiong, Zihuang Wu, Lei Zhang, Lei Lu, Ming Li, Guanbin Li | Paper / Code |
| Develops an improved high-resolution baseline for adapting SAM2 to downstream segmentation tasks. | |||||
| 2026 | IEEE TCSVT | MGD-SAM2 | Multi-view Guided Detail-enhanced SAM 2 for High-Resolution Segmentation | Haoran Shen, Peixian Zhuang, Jiahao Kou, Yuxin Zeng, Haoying Xu, Jiangyun Li | Paper / Code |
| Enhances SAM2 with multi-view guidance and detail enhancement for high-resolution class-agnostic segmentation. | |||||
| 2025 | ICME | SU-SAM | A Simple Unified Framework for Adapting SAM in Underperformed Scenes | Yiran Song, Qianyu Zhou, Xuequan Lu, Zhiwen Shao, Lizhuang Ma | Paper / Code |
| Proposes a simple unified framework to adapt SAM for challenging scenes where the original model underperforms. |
2026
| Year | Pub. | ποΈ Network | Title | Author | Links |
|---|---|---|---|---|---|
| 2026 | CVPR | FlowDIS | Language-Guided Dichotomous Image Segmentation with Flow Matching | Andranik Sargsyan, Shant Navasardyan | Paper / Project / Code |
| Builds on flow matching to transport the image distribution to the corresponding mask distribution, with optional text-prompt guidance via the Position-Aware Instance Pairing (PAIP) training strategy. | |||||
| 2026 | Sensors | PMG-SAM | Boosting Auto-Segmentation of SAM with Pre-Mask Guidance | Jixue Gao, Xiaoyan Jiang, Anjie Wang, Yongbin Gao, Zhijun Fang, Michael S. Lew | Paper |
| Introduces pre-mask guidance mechanism to boost SAM's auto-segmentation capability for better initial predictions. |
2025
| Year | Pub. | ποΈ Network | Title | Author | Links |
|---|---|---|---|---|---|
| 2025 | ICCV | LawDIS | Language-Window-based Controllable DIS | Xinyu Yan, Meijun Sun, Ge-Peng Ji, Fahad Shahbaz Khan, Salman Khan, Deng-Ping Fan | Paper / Code |
| Enables controllable DIS through language and window-based interaction for flexible user control. | |||||
| 2025 | ICLR | OrderIS | Order-aware Interactive Segmentation | Bin Wang, Anwesa Choudhuri, Meng Zheng, Zhongpai Gao, Benjamin Planche, Andong Deng, Qin Liu, Terrence Chen, Ulas Bagci, Ziyan Wu | Paper |
| Incorporates order-aware learning into interactive segmentation to model the sequential nature of user interactions. | |||||
| 2025 | ICME | DIS-SAM | Promoting SAM towards Highly Accurate DIS | Xianjie Liu, Keren Fu, Yao Jiang, Qijun Zhao | Paper / Code |
| Adapts SAM specifically for high-precision DIS with enhanced boundary awareness and detail preservation. | |||||
| 2025 | ESWA | EG-SAM | An Edge-Guided SAM for Effective Complex Object Segmentation | Longyi Chen, Xiandong Wang, Fengqin Yao, Mingchen Song, Jiaheng Zhang, Shengke Wang | Paper / Code |
| Integrates edge guidance into SAM for more effective segmentation of complex objects with intricate boundaries. |
2024
| Year | Pub. | ποΈ Network | Title | Author | Links |
|---|---|---|---|---|---|
| 2024 | ACM MM | PI-SAM | Segment Anything with Precise Interaction | Mengzhen Liu, Mengyu Wang, Henghui Ding, Yilong Xu, Yao Zhao, Yunchao Wei | Paper |
| Improves SAM's interaction mechanism for more precise user control and better segmentation accuracy. | |||||
| 2024 | CVPR | SegNext | Rethinking Interactive Image Segmentation with Low Latency High Quality | Qin Liu, Jaemin Cho, Mohit Bansal, Marc Niethammer | Paper / Code |
| Rethinks the trade-off between latency and quality in interactive segmentation for practical applications. | |||||
| 2024 | ECCV | CAT-SAM | Conditional Tuning for Few-Shot Adaptation of SAM | Aoran Xiao, Weihao Xuan, Heli Qi, Yun Xing, Ruijie Ren, Xiaoqin Zhang, Ling Shao, Shijian Lu | Paper / Code |
| Proposes conditional tuning approach for few-shot adaptation of SAM to new domains with limited samples. | |||||
| 2024 | ICCV | BA-SAM | Scalable Bias-Mode Attention Mask for SAM | Yiran Song, Qianyu Zhou, Xiangtai Li, Deng-Ping Fan, Xuequan Lu, Lizhuang Ma | Paper |
| Introduces bias-mode attention mask to improve SAM's segmentation quality with scalable design. | |||||
| 2024 | ECCV | OVSAM | Mamba or RWKV: Exploring High-Quality and High-Efficiency SAM | Haobo Yuan, Xiangtai Li, Lu Qi, Tao Zhang, Ming-Hsuan Yang, Shuicheng Yan, Chen Change Loy | Paper / Code |
| Explores Mamba and RWKV architectures as alternatives to transformers for efficient high-quality segmentation. | |||||
| 2024 | ICME | PA-SAM | Prompt Adapter SAM for High-Quality Image Segmentation | Zhaozhi Xie, Bochen Guan, Weihao Jiang, Muyang Yi, Yue Ding, Hongtao Lu, Lei Zhang | Paper / Code |
| Designs a prompt adapter to enhance SAM's capability for high-quality image segmentation tasks. | |||||
| 2024 | PRICAI | BSRNet | Prior Mask-Guided Highly Accurate DIS | Shanfeng Zhou, Bo Yuan, Keren Fu, Hailun Zhang, Qijun Zhao | Paper / Code |
| Utilizes prior mask guidance to achieve highly accurate dichotomous image segmentation. |
2023
| Year | Pub. | ποΈ Network | Title | Author | Links |
|---|---|---|---|---|---|
| 2023 | NeurIPS | SAM-HQ | Segment Anything in High Quality | Lei Ke, Mingqiao Ye, Martin Danelljan, Yifan Liu, Yu-Wing Tai, Chi-Keung Tang, Fisher Yu | Paper / Code |
| π Enhances SAM with high-quality output for finer boundaries and more accurate segmentation masks. | |||||
| 2023 | NeurIPS | SegRefiner | Towards Model-Agnostic Segmentation Refinement with Discrete Diffusion | Mengyu Wang, Henghui Ding, Jun Hao Liew, Jiajun Liu, Yao Zhao, Yunchao Wei | Paper / Code |
| Proposes a model-agnostic refinement approach using discrete diffusion to improve segmentation quality. |
β Star History
If you find this repository useful, please consider giving it a star β!
π License
This project is licensed under the MIT License - see the LICENSE file for details.
Made with β€οΈ for the DIS community
Similar repos
Contributors
Tennine2077
70 commits
AndranikSargsyan
1 commits