{"id":86607,"url":"https://github.com/KwanWaiPang/Awesome-Diffusion-based-SLAM","name":"Awesome-Diffusion-based-SLAM","description":"Paper Survey for Diffusion-based SLAM","projects_count":39,"last_synced_at":"2026-09-01T17:00:25.626Z","repository":{"id":284097138,"uuid":"951748664","full_name":"KwanWaiPang/Awesome-Diffusion-based-SLAM","owner":"KwanWaiPang","description":"Paper Survey for Diffusion-based SLAM","archived":false,"fork":false,"pushed_at":"2026-05-02T07:35:51.000Z","size":87,"stargazers_count":35,"open_issues_count":0,"forks_count":3,"subscribers_count":2,"default_branch":"main","last_synced_at":"2026-08-13T02:07:35.831Z","etag":null,"topics":["awesome","deep-learning","depth-estimation","diffusion-models","pose-estimation","slam"],"latest_commit_sha":null,"homepage":"https://kwanwaipang.github.io/Diffusion_SLAM/","language":null,"has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":null,"status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/KwanWaiPang.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":null,"code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null,"governance":null,"roadmap":null,"authors":null,"dei":null,"publiccode":null,"codemeta":null,"zenodo":null}},"created_at":"2025-03-20T07:20:21.000Z","updated_at":"2026-07-28T07:50:25.000Z","dependencies_parsed_at":"2025-04-25T17:01:52.512Z","dependency_job_id":"1ef07ad8-a059-4189-9154-1dc8601d3750","html_url":"https://github.com/KwanWaiPang/Awesome-Diffusion-based-SLAM","commit_stats":null,"previous_names":["kwanwaipang/awesome-diffusion-based-slam"],"tags_count":0,"template":false,"template_full_name":null,"purl":"pkg:github/KwanWaiPang/Awesome-Diffusion-based-SLAM","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/KwanWaiPang%2FAwesome-Diffusion-based-SLAM","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/KwanWaiPang%2FAwesome-Diffusion-based-SLAM/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/KwanWaiPang%2FAwesome-Diffusion-based-SLAM/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/KwanWaiPang%2FAwesome-Diffusion-based-SLAM/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/KwanWaiPang","download_url":"https://codeload.github.com/KwanWaiPang/Awesome-Diffusion-based-SLAM/tar.gz/refs/heads/main","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/KwanWaiPang%2FAwesome-Diffusion-based-SLAM/sbom","scorecard":null,"host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":286080680,"owners_count":37018708,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2026-08-22T15:14:58.755Z","status":"online","status_checked_at":"2026-09-01T02:00:05.433Z","response_time":53,"last_error":null,"robots_txt_status":"success","robots_txt_updated_at":"2025-07-24T06:49:26.215Z","robots_txt_url":"https://github.com/robots.txt","online":true,"can_crawl_api":true,"host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"created_at":"2025-03-25T00:01:44.103Z","updated_at":"2026-09-01T17:00:25.626Z","primary_language":null,"list_of_lists":false,"displayable":true,"categories":["Other Resource","Depth Estimation","Matching","Pose Estimation"],"sub_categories":[],"readme":"\u003cp align=\"center\"\u003e\n  \u003ch1 align=\"center\"\u003e\n  Awesome Diffusion-based SLAM\n  \u003c/h1\u003e\n\u003c/p\u003e\n\nThis repository contains a curated list of resources addressing SLAM-related tasks employing the Diffusion Model, including view/feature correspondences, depth estimation, 3D reconstruction, pose estimation, etc.\n\nIf you find some ignored papers, **feel free to [*create pull requests*](https://github.com/KwanWaiPang/Awesome-Transformer-based-SLAM/blob/pdf/How-to-PR.md), or [*open issues*](https://github.com/KwanWaiPang/Awesome-Diffusion-based-SLAM/issues/new)**. \n\nContributions in any form to make this list more comprehensive are welcome.\n\nIf you find this repository useful, a simple star should be the best affirmation. 😊\n\nFeel free to share this list with others!\n\n# Overview\n\n- [Matching](#Matching)\n- [Depth Estimation](#Depth-Estimation)\n- [Pose Estimation](#Pose-Estimation)\n- [Other Resource](#Other-Resource)\n\n\n## Matching\nor data association, or correspondence\n\n\u003c!-- |---|`arXiv`|---|---|---| --\u003e\n\u003c!-- [![Github stars](https://img.shields.io/github/stars/***.svg)]() --\u003e\n\n| Year | Venue | Paper Title | Repository | Note |\n|:----:|:-----:| ----------- |:----------:|:----:|\n|2025|`arXiv`|[MATCHA: Towards Matching Anything](https://arxiv.org/pdf/2501.14945)|---|SD+DINOv2|\n|2024|`CVPR`|[Sd4match: Learning to prompt stable diffusion model for semantic matching](https://openaccess.thecvf.com/content/CVPR2024/papers/Li_SD4Match_Learning_to_Prompt_Stable_Diffusion_Model_for_Semantic_Matching_CVPR_2024_paper.pdf)|[![Github stars](https://img.shields.io/github/stars/ActiveVisionLab/SD4Match.svg)](https://github.com/ActiveVisionLab/SD4Match)|[website](https://sd4match.active.vision/)|\n|2023|`NIPS`|[Emergent correspondence from image diffusion](https://proceedings.neurips.cc/paper_files/paper/2023/file/0503f5dce343a1d06d16ba103dd52db1-Paper-Conference.pdf)|[![Github stars](https://img.shields.io/github/stars/Tsingularity/dift.svg)](https://github.com/Tsingularity/dift)|[website](https://diffusionfeatures.github.io/)\u003cbr\u003eDIFT|\n|2023|`NIPS`|[A tale of two features: Stable diffusion complements dino for zero-shot semantic correspondence](https://proceedings.neurips.cc/paper_files/paper/2023/file/8e9bdc23f169a05ea9b72ccef4574551-Paper-Conference.pdf)|[![Github stars](https://img.shields.io/github/stars/Junyi42/sd-dino.svg)](https://github.com/Junyi42/sd-dino)|[website](https://sd-complements-dino.github.io/)\u003cbr\u003eSD+DINO|\n|2023|`NIPS`|[Diffusion hyperfeatures: Searching through time and space for semantic correspondence](https://proceedings.neurips.cc/paper_files/paper/2023/file/942032b61720a3fd64897efe46237c81-Paper-Conference.pdf)|[![Github stars](https://img.shields.io/github/stars/diffusion-hyperfeatures/diffusion_hyperfeatures.svg)](https://github.com/diffusion-hyperfeatures/diffusion_hyperfeatures)|[website](https://diffusion-hyperfeatures.github.io/)\u003cbr\u003eDHF|\n\n\n## Depth Estimation\n\nor 3D reconstruction\n\n\u003c!-- |---|`arXiv`|---|---|---| --\u003e\n\u003c!-- [![Github stars](https://img.shields.io/github/stars/***.svg)]() --\u003e\n\n| Year | Venue | Paper Title | Repository | Note |\n|:----:|:-----:| ----------- |:----------:|:----:|\n|2026|`ICLR`|[NOVA3R: Non-pixel-aligned Visual Transformer for Amodal 3D Reconstruction](https://arxiv.org/pdf/2603.04179)|[![Github stars](https://img.shields.io/github/stars/wrchen530/nova3r.svg)](https://github.com/wrchen530/nova3r)|[website](https://wrchen530.github.io/nova3r/) \u003cbr\u003e VGGT等模型是基于pixel alignment的，NOVA3R则是不再让几何必须依附于输入图像中的像素位置（非像素对齐），而是直接学习一个全局一致、视图无关的场景表示，最后再通过3D 扩散解码器恢复全局完整点云；引入了一个基于扩散的三维自编码器，将完整点云压缩为紧凑的潜在表示，再通过流匹配训练解码回原始点云空间，从而处理无序点集中的匹配歧义。在VGGT的预训练图像编码器上引入可学习的 scene tokens，用于聚合任意数量输入视图的信息，并将其映射到统一的场景潜在空间。|\n|2025|`arXiv`|[SpatialCrafter: Unleashing the Imagination of Video Diffusion Models for Scene Reconstruction from Limited Observations](https://arxiv.org/pdf/2505.11992)|---|---|\n|2025|`arXiv`|[Geo4D: Leveraging Video Generators for Geometric 4D Scene Reconstruction](https://arxiv.org/pdf/2504.07961)|[![Github stars](https://img.shields.io/github/stars/jzr99/Geo4D.svg)](https://github.com/jzr99/Geo4D)|[website](https://geo4d.github.io/)\u003cbr\u003evideo generator + MonST3R/DUSt3R|\n|2025|`arXiv`|[Scene Splatter: Momentum 3D Scene Generation from Single Image with Video Diffusion Model](https://arxiv.org/pdf/2504.02764)|---|[website](https://shengjun-zhang.github.io/SceneSplatter/)|\n|2025|`arXiv`|[Can Video Diffusion Model Reconstruct 4D Geometry ?](https://arxiv.org/pdf/2503.21082)|---|[website](https://wayne-mai.github.io/publication/sora3r_arxiv_2025/)\u003cbr\u003eSora3R|\n|2025|`arXiv`|[Bolt3D: Generating 3D Scenes in Seconds](https://arxiv.org/pdf/2503.14445)|---|[website](https://szymanowiczs.github.io/bolt3d)|\n|2025|`arXiv`|[Stable Virtual Camera: Generative View Synthesis with Diffusion Models](https://arxiv.org/pdf/2503.14489)|[![Github stars](https://img.shields.io/github/stars/Stability-AI/stable-virtual-camera.svg)](https://github.com/Stability-AI/stable-virtual-camera) |---|\n|2025|`CVPR`|[Free360: Layered Gaussian Splatting for Unbounded 360-Degree View Synthesis from Extremely Sparse and Unposed Views](https://arxiv.org/pdf/2503.24382)|[![Github stars](https://img.shields.io/github/stars/chobao/Free360.svg)](https://github.com/chobao/Free360)|[website](https://zju3dv.github.io/free360/)| \n|2025|`CVPR`|[GenFusion: Closing the Loop between Reconstruction and Generation via Videos](https://arxiv.org/pdf/2503.21219)|[![Github stars](https://img.shields.io/github/stars/Inception3D/GenFusion.svg)](https://github.com/Inception3D/GenFusion)|[website](https://genfusion.sibowu.com/)|\n|2025|`CVPR`|[Learning temporally consistent video depth from video diffusion priors](https://arxiv.org/pdf/2406.01493)|[![Github stars](https://img.shields.io/github/stars/jiahao-shao1/ChronoDepth.svg)](https://github.com/jiahao-shao1/ChronoDepth)|[website](https://xdimlab.github.io/ChronoDepth/)| \n|2025|`CVPR`|[Depthcrafter: Generating consistent long depth sequences for open-world videos](https://arxiv.org/pdf/2409.02095)|[![Github stars](https://img.shields.io/github/stars/Tencent/DepthCrafter.svg)](https://github.com/Tencent/DepthCrafter)|[website](https://depthcrafter.github.io/)|\n|2025|`CVPR`|[Multi-view Reconstruction via SfM-guided Monocular Depth Estimation](https://arxiv.org/pdf/2503.14483)|[![Github stars](https://img.shields.io/github/stars/zju3dv/Murre.svg)](https://github.com/zju3dv/Murre)|[website](https://zju3dv.github.io/murre/)\u003cbr\u003eMurre|\n|2025|`CVPR`|[Difix3D+: Improving 3D Reconstructions with Single-Step Diffusion Models](https://arxiv.org/pdf/2503.01774?)|---|[website](https://research.nvidia.com/labs/toronto-ai/difix3d/)|\n|2025|`CVPR`|[Align3r: Aligned monocular depth estimation for dynamic videos](https://arxiv.org/pdf/2412.03079)|[![Github stars](https://img.shields.io/github/stars/jiah-cloud/Align3R.svg)](https://github.com/jiah-cloud/Align3R)|---|\n|2024|`NIPS`|[Cat3d: Create anything in 3d with multi-view diffusion models](https://arxiv.org/pdf/2405.10314)|---|[website](https://cat3d.github.io/)|\n|2024|`CVPR`|[Repurposing Diffusion-Based Image Generators for Monocular Depth Estimation](https://openaccess.thecvf.com/content/CVPR2024/papers/Ke_Repurposing_Diffusion-Based_Image_Generators_for_Monocular_Depth_Estimation_CVPR_2024_paper.pdf)| [![Github stars](https://img.shields.io/github/stars/prs-eth/marigold.svg)](https://github.com/prs-eth/marigold)|[website](https://marigoldmonodepth.github.io/)|\n|2024|`ECCV`|[Diffusiondepth: Diffusion denoising approach for monocular depth estimation](https://arxiv.org/pdf/2303.05021)|[![Github stars](https://img.shields.io/github/stars/duanyiqun/DiffusionDepth.svg)](https://github.com/duanyiqun/DiffusionDepth)|[website](https://igl-hkust.github.io/Align3R.github.io/)|\n|2024|`arXiv`|[World-consistent Video Diffusion with Explicit 3D Modeling](https://arxiv.org/pdf/2412.01821)|---|[website](https://zqh0253.github.io/wvd/)| \n|2023|`NIPS`|[The surprising effectiveness of diffusion models for optical flow and monocular depth estimation](https://proceedings.neurips.cc/paper_files/paper/2023/file/7c119415672ae2186e17d492e1d5da2f-Paper-Conference.pdf)|---|[website](https://diffusion-vision.github.io/)|\n|2023|`arXiv`|[Monocular depth estimation using diffusion models](https://arxiv.org/pdf/2302.14816)|---|[website](https://depth-gen.github.io/)| \n|2023|`arXiv`|[Mvdream: Multi-view diffusion for 3d generation](https://arxiv.org/pdf/2308.16512)|[![Github stars](https://img.shields.io/github/stars/bytedance/MVDream.svg)](https://github.com/bytedance/MVDream)|[website](https://mv-dream.github.io/)|\n\n## Pose Estimation\n\n\u003c!-- |---|`arXiv`|---|---|---| --\u003e\n\u003c!-- [![Github stars](https://img.shields.io/github/stars/***.svg)]() --\u003e\n\n| Year | Venue | Paper Title | Repository | Note |\n|:----:|:-----:| ----------- |:----------:|:----:|\n|2025|`arXiv`|[GCRayDiffusion: Pose-Free Surface Reconstruction via Geometric Consistent Ray Diffusion](https://arxiv.org/pdf/2503.22349)|---|---|\n|2023|`ICCV`|[Posediffusion: Solving pose estimation via diffusion-aided bundle adjustment](https://openaccess.thecvf.com/content/ICCV2023/papers/Wang_PoseDiffusion_Solving_Pose_Estimation_via_Diffusion-aided_Bundle_Adjustment_ICCV_2023_paper.pdf)|[![Github stars](https://img.shields.io/github/stars/facebookresearch/PoseDiffusion.svg)](https://github.com/facebookresearch/PoseDiffusion)|[website](https://posediffusion.github.io/)|\n\n\n## Other Resource\n\n* Survey for Learning-based VO,VIO,IO：[Paper List](https://github.com/KwanWaiPang/Awesome-Learning-based-VO-VIO)\n* Survey for Transformer-based SLAM：[Paper List](https://github.com/KwanWaiPang/Awesome-Transformer-based-SLAM)\n* [Awesome-Diffusion-Models](https://github.com/diff-usion/Awesome-Diffusion-Models)\n* Some basic paper for Diffusion Model:\n\n\u003c!-- |---|`arXiv`|---|---|---| --\u003e\n\u003c!-- [![Github stars](https://img.shields.io/github/stars/***.svg)]() --\u003e\n\n| Year | Venue | Paper Title | Repository | Note |\n|:----:|:-----:| ----------- |:----------:|:----:|\n|2022|`CVPR`|[High-resolution image synthesis with latent diffusion models](https://openaccess.thecvf.com/content/CVPR2022/papers/Rombach_High-Resolution_Image_Synthesis_With_Latent_Diffusion_Models_CVPR_2022_paper.pdf)|[![Github stars](https://img.shields.io/github/stars/CompVis/latent-diffusion.svg)](https://github.com/CompVis/latent-diffusion)|stable diffusion|\n|2021|`NIPS`|[Diffusion models beat gans on image synthesis](https://proceedings.neurips.cc/paper_files/paper/2021/file/49ad23d1ec9fa4bd8d77d02681df5cfa-Paper.pdf)|---|Ablated Diffusion Model(ADM)|\n|2020|`ICLR`|[Denoising diffusion implicit models](https://arxiv.org/pdf/2010.02502)|---|DDIM|\n|2020|`NIPS`|[Denoising diffusion probabilistic models](https://proceedings.neurips.cc/paper/2020/file/4c5bcfec8584af0d967f1ab10179ca4b-Paper.pdf)|[![Github stars](https://img.shields.io/github/stars/hojonathanho/diffusion.svg)](https://github.com/hojonathanho/diffusion)|DDPM|\n\n\n\n\n\n","projects_url":"https://awesome.ecosyste.ms/api/v1/lists/kwanwaipang%2Fawesome-diffusion-based-slam/projects"}