Ecosyste.ms: Awesome
An open API service indexing awesome lists of open source software.
https://github.com/younader/vesuvius-grandprize-winner
https://github.com/younader/vesuvius-grandprize-winner
Last synced: about 4 hours ago
JSON representation
- Host: GitHub
- URL: https://github.com/younader/vesuvius-grandprize-winner
- Owner: younader
- License: mit
- Created: 2024-02-05T09:11:01.000Z (9 months ago)
- Default Branch: main
- Last Pushed: 2024-10-02T20:19:16.000Z (about 1 month ago)
- Last Synced: 2024-10-14T07:31:57.198Z (25 days ago)
- Language: Python
- Size: 42.8 MB
- Stars: 578
- Watchers: 12
- Forks: 34
- Open Issues: 0
-
Metadata Files:
- Readme: readme.md
- License: LICENSE.txt
Awesome Lists containing this project
README
# Vesuvius Grandprize Winning Solution
The Repository contains the First Place Vesuvius Grand Prize solution.
This repository is part of the First Place Grand Prize Submission to the Vesuvius Challenge 2023 from Youssef Nader, Luke Farritor and Julian Schilliger.
## Automatic Segmentation 
Check out the automatic segmentation pipeline ThaumatoAnakalyptor of our winning Grand Prize submission by Julian Schilliger.
[ThaumatoAnakalyptor](https://github.com/schillij95/ThaumatoAnakalyptor/tree/main) performs in full 3D and is also capable of segmenting in very mushy and twisted scroll regions.
# Ink Detection Overview
:
Our final canonical model was a timesformer small architecture with divided space-time attention.
The dataset underwent expansion and cleaning rounds to increase accuracy of the labels and become as accurate as possible, approximately 15 rounds were performed between the first letters and final solution.
Our solution also consisted of 2 other architectures, Resnet3D-101 with pretrained weights, I3D with non-local block and maxpooling.
Our implementation uses `torch`, `torch-lightning`,the [`timesformer-pytorch`](https://github.com/lucidrains/TimeSformer-pytorch) and [`3D-ResNets-PyTorch`](https://github.com/kenshohara/3D-ResNets-PyTorch/blob/master/models/resnet.py).
# 🚀 Get Started
EASY: build the docker image:
```bash
docker build -t youssef_gp .
docker run --gpus all --shm-size=150g -it -v :/workspace/train_scrolls youssef_gp
```
Then to train:
```bash
python train_timesformer_og.py
```
Or to run inference with the already trained model:
```bash
python inference_timesformer.py --model_path timesformer_weights.ckpt --segment_path train_scrolls --segment_id 20231005123336
```
Important note: to install the ink labels and training data inside the docker image, run:
```bash
#to download the segments from the server
./download.sh
#propagates the inklabels into the respective segment folders for training
python prepare.py
```
You can find the weights of the canonical timesformer uploaded [here](https://drive.google.com/drive/folders/1rn3GMOvtJRMBHOxVhWFVSY6IVI6xUnYp?usp=sharing)
# Example Inference
To run inference of timesformer:
```bash
python inference_timesformer.py --segment_id 20231210121321 20231221180251 --segment_path $(pwd)/train_scrolls --model_path timesformer_weights.ckpt
```
The optional parameter ```--out_path``` can be used to specify the output path of the predictions.