GeoViG: Geometry-Aware Graph Reasoning for Mobile Vision Tasks
Model Variants
πΌοΈ Image Classification β ImageNet-1K
| Model | Params (M) | MACs (G) | Top-1 Acc (%) | Checkpoint (pth) |
|---|---|---|---|---|
| GeoViG-Ti | 3.5 | 0.9 | 75.2 | Download |
| GeoViG-S | 5.0 | 1.2 | 77.5 | Download |
| GeoViG-M | 10.3 | 2.2 | 80.7 | Download |
| GeoViG-B | 19.7 | 4.5 | 82.4 | Download |
Trained for 300 epochs on 8Γ NVIDIA A100 GPUs, batch size 1024, AdamW optimizer (lr=5e-4, weight decay=0.05), cosine schedule with 5-epoch warmup. Augmentations: RandAugment, Mixup (p=0.8), CutMix (p=1.0).
For CoreML checkpoint, please check: https://huggingface.co/OmarAlasqa/GeoViG/tree/main/CoreML For IPA chechpoints, please check: https://huggingface.co/OmarAlasqa/GeoViG/tree/main/IPA
π¦ Object Detection & Instance Segmentation β MS COCO 2017
Backbone used with Mask R-CNN, 1Γ schedule (12 epochs), pretrained on ImageNet-1K.
| Backbone | Params (M) | Box AP | Box APβ β | Box APββ | Mask AP | Mask APβ β | Mask APββ | Checkpoint |
|---|---|---|---|---|---|---|---|---|
| GeoViG-M | 10.3 | 40.7 | 62.4 | 44.1 | 37.7 | 59.6 | 40.5 | Download |
| GeoViG-B | 19.7 | 42.5 | 64.0 | 46.8 | 38.9 | 61.2 | 41.7 | Download |
π₯ Medical Image Segmentation
Kvasir-SEG β Polyp Segmentation
| Model | Params (M) | mAP | Dice β | IoU β | Hausdorff Dist β | Checkpoint |
|---|---|---|---|---|---|---|
| GeoViG-M | 29.57 | 0.990 | 0.945 | 0.909 | 12.94 | Download |
Data Science Bowl 2018 β Nuclei Segmentation
| Model | Params (M) | mAP | Dice β | IoU β | Hausdorff Dist β | Checkpoint |
|---|---|---|---|---|---|---|
| GeoViG-M | 29.57 | 0.859 | 0.908 | 0.839 | 5.19 | Download |
π Usage
Please check the github repo: https://github.com/OmarAlsaqa/GeoViG
Citation
If you use GeoViG in your research, please cite:
@article{alsaqa2026geovig,
title = {GeoViG: Geometry-Aware Graph Reasoning for Mobile Vision Tasks in Natural and Medical Images},
author = {Alsaqa, Omar and Mohammed, Emad and Aleem, Saiqa},
journal = {Under Review at IEEE EMBC},
year = {2026}
}
Acknowledgements
This work builds upon MobileViG and uses the MMDetection framework for detection and segmentation experiments. Training was performed on the Compute Canada A100 cluster.
License
This project is released under the Apache 2.0 License.
- Downloads last month
- 7