Foreseeing the Invisible: Amodal Reconstruction of Leaf Fossil Images
Quick Answer
AmodalDINO is a novel multi-head dense-prediction model that achieves 95.0% Dice and 90.5% IoU in amodal reconstruction of leaf fossils, predicting visible and complete leaf structures from a single RGB image without prior visible masks.
Quick Take
The model, trained on synthetic images, runs offline in a browser with 4-bit weights, demonstrating practical applications in paleobotany.
Key Points
- AmodalDINO predicts four masks: visible leaf, complete leaf, main vein, and fine veins.
- The model adapts DINOv3 ViT-L/16 with fine-tuning and auxiliary venation heads.
- Achieved 85.05 full mIoU on KINS and 80.90 on COCOA-cls with two heads.
- Runs offline in a browser with 4-bit quantization, matching original model performance.
- Includes ruler-based calibration for estimating surface area of fossil leaves.
Paper Resources
📖 Reader Mode
~2 min readAbstract:Fossil leaves are rarely preserved whole -- sedimentary rock hides, breaks, and erodes the lamina, yet paleobotany depends on the complete shape and outline of the leaf. We cast the recovery of the missing tissue as amodal reconstruction and present AmodalDINO, a multi-head dense-prediction model that predicts four masks from a single RGB image: visible leaf, amodal complete leaf, amodal main vein, and fine veins. Unlike essentially all prior amodal work, AmodalDINO is given no visible mask. It predicts the visible and amodal regions jointly, so it needs no upstream instance segmenter at runtime. Two simple but effective changes adapt the model to the amodal segmentation task: fully fine-tune a DINOv3 ViT-L/16 at a small learning rate instead of freezing it, and attach auxiliary venation heads alongside the leaf heads. These two changes enable the model to learn the structural shape prior of leaves. Trained only on synthetic leaf fossil images, AmodalDINO reaches 95.0% Dice / 90.5% IoU on the validation set and transfers well to real fossil specimens. Stripped to two heads, the same recipe can run on two benchmark datasets, reaching 85.05 full mIoU / 66.65 occluded mIoU on KINS and 80.90 / 38.15 on COCOA-cls. The model is also practical: by quantizing to 4-bit weights, it runs entirely offline in a browser, matching the original model with an IoU of 0.910. We also add ruler-based calibration to estimate surface area, and a generative visualization of living leaves on local devices.
| Comments: | 12 pages, 10 figures |
| Subjects: | Computer Vision and Pattern Recognition (cs.CV) |
| Cite as: | arXiv:2608.04423 [cs.CV] |
| (or arXiv:2608.04423v1 [cs.CV] for this version) | |
| https://doi.org/10.48550/arXiv.2608.04423 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Liuxiang Yue [view email]
[v1]
Wed, 5 Aug 2026 04:07:15 UTC (2,689 KB)
— Originally published at arxiv.org
Want this in your inbox every morning?
Daily brief at your local 8am — bilingual EN/中文, free.
More from arXiv cs.CV
See more →ProMoE-FL: Prototype-conditioned Mixture of Experts for Multimodal Federated Learning with Missing Modalities
ProMoE-FL introduces a Prototype-conditioned Mixture-of-Experts framework for multimodal federated learning, effectively addressing missing modalities. It outperforms existing methods on four chest X-ray datasets, demonstrating superior feature synthesis capabilities in both homogeneous and heterogeneous settings.