[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"project-71989":3},{"id":4,"name":5,"fullName":6,"owner":7,"repo":5,"description":8,"homepage":9,"htmlUrl":10,"language":11,"languages":10,"totalLinesOfCode":10,"stars":12,"forks":13,"watchers":14,"openIssues":15,"contributorsCount":16,"subscribersCount":16,"size":16,"stars1d":17,"stars7d":18,"stars30d":19,"stars90d":16,"forks30d":16,"starsTrendScore":20,"compositeScore":21,"rankGlobal":10,"rankLanguage":10,"license":22,"archived":23,"fork":23,"defaultBranch":24,"hasWiki":23,"hasPages":25,"topics":26,"createdAt":10,"pushedAt":10,"updatedAt":27,"readmeContent":28,"aiSummary":29,"trendingCount":16,"starSnapshotCount":16,"syncStatus":30,"lastSyncTime":31,"discoverSource":32},71989,"ml-sharp","apple\u002Fml-sharp","apple","Sharp Monocular View Synthesis in Less Than a Second","https:\u002F\u002Fapple.github.io\u002Fml-sharp\u002F",null,"Python",8505,617,46,53,0,30,56,181,90,39.37,"Other",false,"main",true,[],"2026-06-12 02:02:57","# Sharp Monocular View Synthesis in Less Than a Second\n\n[![Project Page](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FProject-Page-green)](https:\u002F\u002Fapple.github.io\u002Fml-sharp\u002F)\n[![arXiv](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FarXiv-2512.10685-b31b1b.svg)](https:\u002F\u002Farxiv.org\u002Fabs\u002F2512.10685)\n\nThis software project accompanies the research paper: _Sharp Monocular View Synthesis in Less Than a Second_\nby _Lars Mescheder, Wei Dong, Shiwei Li, Xuyang Bai, Marcel Santos, Peiyun Hu, Bruno Lecouat, Mingmin Zhen, Amaël Delaunoy,\nTian Fang, Yanghai Tsin, Stephan Richter and Vladlen Koltun_.\n\n![](data\u002Fteaser.jpg)\n\nWe present SHARP, an approach to photorealistic view synthesis from a single image. Given a single photograph, SHARP regresses the parameters of a 3D Gaussian representation of the depicted scene. This is done in less than a second on a standard GPU via a single feedforward pass through a neural network. The 3D Gaussian representation produced by SHARP can then be rendered in real time, yielding high-resolution photorealistic images for nearby views. The representation is metric, with absolute scale, supporting metric camera movements. Experimental results demonstrate that SHARP delivers robust zero-shot generalization across datasets. It sets a new state of the art on multiple datasets, reducing LPIPS by 25–34% and DISTS by 21–43% versus the best prior model, while lowering the synthesis time by three orders of magnitude.\n\n## Getting started\n\nWe recommend to first create a python environment:\n\n```\nconda create -n sharp python=3.13\n```\n\nAfterwards, you can install the project using\n\n```\npip install -r requirements.txt\n```\n\nTo test the installation, run\n\n```\nsharp --help\n```\n\n## Using the CLI\n\nTo run prediction:\n\n```\nsharp predict -i \u002Fpath\u002Fto\u002Finput\u002Fimages -o \u002Fpath\u002Fto\u002Foutput\u002Fgaussians\n```\n\nThe model checkpoint will be downloaded automatically on first run and cached locally at `~\u002F.cache\u002Ftorch\u002Fhub\u002Fcheckpoints\u002F`.\n\nAlternatively, you can download the model directly:\n\n```\nwget https:\u002F\u002Fml-site.cdn-apple.com\u002Fmodels\u002Fsharp\u002Fsharp_2572gikvuh.pt\n```\n\nTo use a manually downloaded checkpoint, specify it with the `-c` flag:\n\n```\nsharp predict -i \u002Fpath\u002Fto\u002Finput\u002Fimages -o \u002Fpath\u002Fto\u002Foutput\u002Fgaussians -c sharp_2572gikvuh.pt\n```\n\nThe results will be 3D gaussian splats (3DGS) in the output folder. The 3DGS `.ply` files are compatible to various public 3DGS renderers. We follow the OpenCV coordinate convention (x right, y down, z forward). The 3DGS scene center is roughly at (0, 0, +z). When dealing with 3rdparty renderers, please scale and rotate to re-center the scene accordingly.\n\n### Rendering trajectories (CUDA GPU only)\n\nAdditionally you can render videos with a camera trajectory. While the gaussians prediction works for all CPU, CUDA, and MPS, rendering videos via the `--render` option currently requires a CUDA GPU. The gsplat renderer takes a while to initialize at the first launch.\n\n```\nsharp predict -i \u002Fpath\u002Fto\u002Finput\u002Fimages -o \u002Fpath\u002Fto\u002Foutput\u002Fgaussians --render\n\n# Or from the intermediate gaussians:\nsharp render -i \u002Fpath\u002Fto\u002Foutput\u002Fgaussians -o \u002Fpath\u002Fto\u002Foutput\u002Frenderings\n```\n\n## Evaluation\n\nPlease refer to the paper for both quantitative and qualitative evaluations.\nAdditionally, please check out this [qualitative examples page](https:\u002F\u002Fapple.github.io\u002Fml-sharp\u002F) containing several video comparisons against related work.\n\n## Citation\n\nIf you find our work useful, please cite the following paper:\n\n```bibtex\n@inproceedings{Sharp2025:arxiv,\n  title      = {Sharp Monocular View Synthesis in Less Than a Second},\n  author     = {Lars Mescheder and Wei Dong and Shiwei Li and Xuyang Bai and Marcel Santos and Peiyun Hu and Bruno Lecouat and Mingmin Zhen and Ama\\\"{e}l Delaunoy and Tian Fang and Yanghai Tsin and Stephan R. Richter and Vladlen Koltun},\n  journal    = {arXiv preprint arXiv:2512.10685},\n  year       = {2025},\n  url        = {https:\u002F\u002Farxiv.org\u002Fabs\u002F2512.10685},\n}\n```\n\n## Acknowledgements\n\nOur codebase is built using multiple opensource contributions, please see [ACKNOWLEDGEMENTS](ACKNOWLEDGEMENTS) for more details.\n\n## License\n\nPlease check out the repository [LICENSE](LICENSE) before using the provided code and\n[LICENSE_MODEL](LICENSE_MODEL) for the released models.\n","SHARP 是一个用于从单张图像快速生成逼真视角合成的软件项目。其核心功能在于通过神经网络的前向传递，在标准GPU上不到一秒的时间内，从给有的单幅照片中回归出场景的3D高斯表示。该3D表示支持实时渲染，能够产生高质量的近景视图，并且具有绝对尺度，允许进行度量级别的相机移动。实验结果显示，SHARP在多个数据集上实现了零样本泛化，相比之前的最佳模型，在LPIPS和DISTS指标上分别降低了25-34%和21-43%，同时大幅减少了合成时间。此工具适用于需要高效、高质量视角合成的应用场景，如虚拟现实、增强现实以及计算机视觉研究等领域。",2,"2026-06-11 03:39:50","high_star"]