{ "id": 487430, "title": "IMC 2024 materials", "url": "/competitions/image-matching-challenge-2024/discussion/487430", "author_name": "Abdo Nasser", "post_date": "2024-03-29T01:57:48.494000", "votes": 19, "comment_count": 0, "views": 0, "content": "

This is a comprehensive guide to begin with the competition, intended for both beginners and advanced individuals. Creating a 3D model of a scene using a bunch of pictures taken from different angles is a tricky problem and requires a strong understanding of structure from motion and computer vision in general.

\n

Good luck to you all in the competition 😀!

\n", "messages": [ { "id": 2721394, "postDate": "2024-03-29T01:57:48.493Z", "content": "

This is a comprehensive guide to begin with the competition, intended for both beginners and advanced individuals. Creating a 3D model of a scene using a bunch of pictures taken from different angles is a tricky problem and requires a strong understanding of structure from motion and computer vision in general.

\n

Good luck to you all in the competition 😀!

\n", "rawMarkdown": "This is a comprehensive guide to begin with the competition, intended for both beginners and advanced individuals. Creating a 3D model of a scene using a bunch of pictures taken from different angles is a tricky problem and requires a strong understanding of structure from motion and computer vision in general.\n\n\nGood luck to you all in the competition 😀!\n\n\n- Youtube :\n 1. [Optical Flow | Structure from Motion | Object Tracking ( Video 7 to 11) ](https://youtube.com/playlist?list=PL2zRqk16wsdoYzrWStffqBAoUY8XdvatV&si=qWTjHGQmhxWvNPsB)\n 2. [3D Computer Vision | Lecture 10 (Part 1): Structure-from-Motion (SfM) and bundle adjustment\n](https://youtu.be/MUadR35FFqk?si=3OVufIByN5FH_JTe)\n 3. [3D Computer Vision | Lecture 10 (Part 2): Structure-from-Motion (SfM) and bundle adjustment](https://youtu.be/j3VSmUCKZDM?si=Cz0rUoI5a4YK80jg)\n 4. [2nd Place Solution for the IMC 2023 video](https://youtu.be/9JpGjpITiDM)\n- Papers :\n 1. [Neural 3D Scene Reconstruction from Multiple 2D Images without 3D Supervision\n](https://arxiv.org/pdf/2306.17643.pdf)\n 2. [Neural 3D Scene Reconstruction with the Manhattan-world Assumption\n](https://arxiv.org/pdf/2205.02836v2.pdf)\n 3. [UniScene: Multi-Camera Unified Pre-training via 3D Scene Reconstruction\n](https://arxiv.org/pdf/2305.18829v4.pdf)\n 4. [Open-source Image-based 3d Reconstruction Pipelines:\nReview, Comparison and Evaluation](https://isprs-archives.copernicus.org/articles/XLII-2-W17/331/2019/isprs-archives-XLII-2-W17-331-2019.pdf)\n\n- Librairies :\n 1. [OpenMVG (Open Multiple View Geometry)](https://github.com/openMVG/openMVG)\n 2. [OpenMVS: open Multi-View Stereo reconstruction library](https://github.com/cdcseacave/openMVS)\n 3. [Meshroom 3D Reconstruction software](https://github.com/alicevision/Meshroom)\n 4. [COLMAP a Structure-from-Motion (SfM) and Multi-View Stereo (MVS) pipeline ](https://colmap.github.io/index.html)\n 5. [Open3D\nA Modern Library for 3D Data Processing](https://www.open3d.org/)\n 6. [OpenCV: Open Source Computer Vision Library](https://github.com/opencv/opencv) \n\n- Others :\n 1. [Awesome 3D reconstruction list of papers](https://github.com/openMVG/awesome_3DReconstruction_list)\n 2. [3D reconstruction from 2D images using openMVG and openMVS](https://medium.com/@popovici.cristina211/3d-reconstruction-from-2d-images-using-openmvg-and-openmvs-b23bc7adb616)\n 3. [3d Reconstruction (from 2d Images)](https://medium.com/@AaronLeeIV/3d-reconstruction-from-2d-images-cb21096631ad)\n\n\n\n\n\n\n\n\n\n\n", "votes": 19 } ], "comments": [], "raw_markdown_by_id": { "2721394": "This is a comprehensive guide to begin with the competition, intended for both beginners and advanced individuals. Creating a 3D model of a scene using a bunch of pictures taken from different angles is a tricky problem and requires a strong understanding of structure from motion and computer vision in general.\n\n\nGood luck to you all in the competition 😀!\n\n\n- Youtube :\n 1. [Optical Flow | Structure from Motion | Object Tracking ( Video 7 to 11) ](https://youtube.com/playlist?list=PL2zRqk16wsdoYzrWStffqBAoUY8XdvatV&si=qWTjHGQmhxWvNPsB)\n 2. [3D Computer Vision | Lecture 10 (Part 1): Structure-from-Motion (SfM) and bundle adjustment\n](https://youtu.be/MUadR35FFqk?si=3OVufIByN5FH_JTe)\n 3. [3D Computer Vision | Lecture 10 (Part 2): Structure-from-Motion (SfM) and bundle adjustment](https://youtu.be/j3VSmUCKZDM?si=Cz0rUoI5a4YK80jg)\n 4. [2nd Place Solution for the IMC 2023 video](https://youtu.be/9JpGjpITiDM)\n- Papers :\n 1. [Neural 3D Scene Reconstruction from Multiple 2D Images without 3D Supervision\n](https://arxiv.org/pdf/2306.17643.pdf)\n 2. [Neural 3D Scene Reconstruction with the Manhattan-world Assumption\n](https://arxiv.org/pdf/2205.02836v2.pdf)\n 3. [UniScene: Multi-Camera Unified Pre-training via 3D Scene Reconstruction\n](https://arxiv.org/pdf/2305.18829v4.pdf)\n 4. [Open-source Image-based 3d Reconstruction Pipelines:\nReview, Comparison and Evaluation](https://isprs-archives.copernicus.org/articles/XLII-2-W17/331/2019/isprs-archives-XLII-2-W17-331-2019.pdf)\n\n- Librairies :\n 1. [OpenMVG (Open Multiple View Geometry)](https://github.com/openMVG/openMVG)\n 2. [OpenMVS: open Multi-View Stereo reconstruction library](https://github.com/cdcseacave/openMVS)\n 3. [Meshroom 3D Reconstruction software](https://github.com/alicevision/Meshroom)\n 4. [COLMAP a Structure-from-Motion (SfM) and Multi-View Stereo (MVS) pipeline ](https://colmap.github.io/index.html)\n 5. [Open3D\nA Modern Library for 3D Data Processing](https://www.open3d.org/)\n 6. [OpenCV: Open Source Computer Vision Library](https://github.com/opencv/opencv) \n\n- Others :\n 1. [Awesome 3D reconstruction list of papers](https://github.com/openMVG/awesome_3DReconstruction_list)\n 2. [3D reconstruction from 2D images using openMVG and openMVS](https://medium.com/@popovici.cristina211/3d-reconstruction-from-2d-images-using-openmvg-and-openmvs-b23bc7adb616)\n 3. [3d Reconstruction (from 2d Images)](https://medium.com/@AaronLeeIV/3d-reconstruction-from-2d-images-cb21096631ad)\n\n\n\n\n\n\n\n\n\n\n" } }