MonoPCC vs vggt
Image-based depth and 3D reconstruction projects compared
MonoPCC estimates monocular depth from endoscopic images using a self-supervised cycle constraint, while vggt predicts multiple kinds of 3D scene information from one or more images. MonoPCC focuses on surgical-video research and dataset evaluations; vggt provides a broader reconstruction workflow with camera, depth, point-map, and track outputs.

MonoPCC: Estimate Monocular Depth in Endoscopic Images
MonoPCC is a PyTorch method for self-supervised monocular depth estimation on endoscopic images, using a photometric-invariant cycle constraint. It is aimed at researchers reproducing or extending depth and pose experiments on surgical-video datasets.

vggt: Reconstruct 3D Scenes from Images
VGGT is a feed-forward vision model that estimates cameras, depth, point maps, and point tracks from one or more images. It suits researchers and developers who need fast 3D scene reconstruction without a traditional multi-stage pipeline.
| MonoPCC | vggt | |
|---|---|---|
| Language | Python | Python |
| License | Apache-2.0 | NOASSERTION |
| Stars | 23 | 14.5k |
| Forks | 1 | 1.6k |
| Last analyzed | Oct 3, 2026 | Oct 3, 2026 |
Key differences
- MonoPCC targets endoscopic depth estimation and reports workflows for SCARED and KITTI, while vggt supports single-view and multi-view scene reconstruction from images.
- MonoPCC uses a photometric-invariant cycle constraint for self-supervised depth estimation; vggt is a feed-forward model that predicts camera parameters, depth, point maps, and tracks.
- MonoPCC includes pose evaluation and a 3D reconstruction demonstration, while vggt offers Python APIs, image and video demos, and COLMAP-format export.
- MonoPCC is licensed under Apache-2.0; vggt lists its license as NOASSERTION, and its checkpoint terms distinguish the original checkpoint from VGGT-1B-Commercial.
- MonoPCC's documented setup uses older software versions and CUDA-based example commands; vggt's pretrained model is large and practical inference is best suited to a CUDA-capable GPU.
- MonoPCC is aimed at researchers studying surgical-scene depth and reconstruction, while vggt also targets developers building general 3D reconstruction and downstream geometry workflows.
Choose MonoPCC if you…
- evaluate self-supervised monocular depth methods on endoscopic video.
- work with SCARED or compare depth-estimation methods on KITTI.
- investigate the PCC strategy, including its use with EndoDAC.
Choose vggt if you…
- need camera, depth, point-map, and track predictions from one or more images.
- want to export reconstructions in COLMAP format for downstream workflows.
- explore feed-forward 3D reconstruction or fine-tune using the included training example.
Compare the numbers side by side, or add more projects
This comparison is generated with AI from the OSRepos analyses of both projects. Always check each project's repository and documentation before choosing.