Merge pull request #285 from zhouzi180/master
Update for Scoliosis1K Dataset
This commit is contained in:
Chengwei Ye
GitHub
@@ -1,34 +1,85 @@
|
||||
# Tutorial for [Scoliosis1K](https://zhouzi180.github.io/Scoliosis1K)
|
||||
## Download the Scoliosis1K Dataset
|
||||
|
||||
## Download the Scoliosis1K dataset
|
||||
Download the dataset from the [link](https://zhouzi180.github.io/Scoliosis1K).
|
||||
decompress these two file by following command:
|
||||
```shell
|
||||
unzip -P password Scoliosis1K-pkl.zip | xargs -n1 tar xzvf
|
||||
```
|
||||
password should be obtained by signing [agreement](https://zhouzi180.github.io/Scoliosis1K/static/resources/Scoliosis1KAgreement.pdf) and sending to email (12331257@mail.sustech.edu.cn)
|
||||
You can download the dataset from the [official website](https://zhouzi180.github.io/Scoliosis1K).
|
||||
The dataset is provided as four compressed files:
|
||||
|
||||
Then you will get Scoliosis1K formatted as:
|
||||
```
|
||||
DATASET_ROOT/
|
||||
00000 (subject)/
|
||||
positive (category)/
|
||||
000-180 (view)/
|
||||
000.pkl (contains all frames)
|
||||
......
|
||||
```
|
||||
## Train the dataset
|
||||
Modify the `dataset_root` in `configs/sconet/sconet_scoliosis1k.yaml`, and then run this command:
|
||||
```shell
|
||||
CUDA_VISIBLE_DEVICES=0,1,2,3 python -m torch.distributed.launch --nproc_per_node=4 opengait/main.py --cfgs configs/sconet/sconet_scoliosis1k.yaml --phase train
|
||||
* `Scoliosis1K-sil-raw.zip`
|
||||
* `Scoliosis1K-sil-pkl.zip`
|
||||
* `Scoliosis1K-pose-raw.zip`
|
||||
* `Scoliosis1K-pose-pkl.zip`
|
||||
|
||||
We recommend using the provided pickle (`.pkl`) files for convenience.
|
||||
Decompress them with the following commands:
|
||||
|
||||
```bash
|
||||
unzip -P <password> Scoliosis1K-sil-pkl.zip
|
||||
unzip -P <password> Scoliosis1K-pose-pkl.zip
|
||||
```
|
||||
|
||||
> **Note**: The \<password\> can be obtained by signing the [release agreement](https://zhouzi180.github.io/Scoliosis1K/static/resources/Scoliosis1k_release_agreement.pdf) and sending it to **[12331257@mail.sustech.edu.cn](mailto:12331257@mail.sustech.edu.cn)**.
|
||||
|
||||
## Process from RAW dataset
|
||||
### Dataset Structure
|
||||
|
||||
After decompression, you will get the following structure:
|
||||
|
||||
### Preprocess the dataset (Optional)
|
||||
Download the raw dataset from the [official link](https://zhouzi180.github.io/Scoliosis1K). You will get two compressed files, i.e. `Scoliosis1K-raw.zip`, and `Scoliosis1K-pkl.zip`.
|
||||
We recommend using our provided pickle files for convenience, or process raw dataset into pickle by this command:
|
||||
```shell
|
||||
python datasets/pretreatment.py --input_path Scoliosis1K_raw --output_path Scoliosis1K-pkl
|
||||
```
|
||||
├── Scoliosis1K-sil-pkl
|
||||
│ ├── 00000 # Identity
|
||||
│ │ ├── Positive # Class
|
||||
│ │ │ ├── 000_180 # View
|
||||
│ │ │ └── 000_180.pkl # Estimated Silhouette (PP-HumanSeg v2)
|
||||
│
|
||||
├── Scoliosis1K-pose-pkl
|
||||
│ ├── 00000 # Identity
|
||||
│ │ ├── Positive # Class
|
||||
│ │ │ ├── 000_180 # View
|
||||
│ │ │ └── 000_180.pkl # Estimated 2D Pose (ViTPose)
|
||||
```
|
||||
|
||||
### Processing from RAW Dataset (optional)
|
||||
|
||||
If you prefer, you can process the raw dataset into `.pkl` format.
|
||||
|
||||
```bash
|
||||
# For silhouette raw data
|
||||
python datasets/pretreatment.py --input_path=<path_to_raw_silhouettes> -output_path=<output_path>
|
||||
|
||||
# For pose raw data
|
||||
python datasets/pretreatment.py --input_path=<path_to_raw_pose> -output_path=<output_path> --pose --dataset=OUMVLP
|
||||
```
|
||||
---
|
||||
|
||||
## Training and Testing
|
||||
|
||||
Before training or testing, modify the `dataset_root` field in
|
||||
`configs/sconet/sconet_scoliosis1k.yaml`.
|
||||
|
||||
Then run the following commands:
|
||||
|
||||
```bash
|
||||
# Training
|
||||
CUDA_VISIBLE_DEVICES=0,1,2,3 \
|
||||
python -m torch.distributed.launch --nproc_per_node=4 \
|
||||
opengait/main.py --cfgs configs/sconet/sconet_scoliosis1k.yaml --phase train --log_to_file
|
||||
|
||||
# Testing
|
||||
CUDA_VISIBLE_DEVICES=0,1,2,3 \
|
||||
python -m torch.distributed.launch --nproc_per_node=4 \
|
||||
opengait/main.py --cfgs configs/sconet/sconet_scoliosis1k.yaml --phase test --log_to_file
|
||||
```
|
||||
|
||||
---
|
||||
|
||||
## Pose-to-Heatmap Conversion
|
||||
|
||||
*From our paper: **Pose as Clinical Prior: Learning Dual Representations for Scoliosis Screening (MICCAI 2025)***
|
||||
|
||||
```bash
|
||||
CUDA_VISIBLE_DEVICES=0,1,2,3 \
|
||||
python -m torch.distributed.launch --nproc_per_node=4 \
|
||||
datasets/pretreatment_heatmap.py \
|
||||
--pose_data_path=<path_to_pose_pkl> \
|
||||
--save_root=<output_path> \
|
||||
--dataset_name=OUMVLP
|
||||
```
|
||||
@@ -5,7 +5,6 @@ from utils import get_msg_mgr, mkdir
|
||||
|
||||
from .metric import mean_iou, cuda_dist, compute_ACC_mAP, evaluate_rank, evaluate_many
|
||||
from .re_rank import re_ranking
|
||||
from sklearn.metrics import confusion_matrix, accuracy_score
|
||||
|
||||
def de_diag(acc, each_angle=False):
|
||||
# Exclude identical-view cases
|
||||
@@ -417,46 +416,49 @@ def evaluate_CCPG(data, dataset, metric='euc'):
|
||||
return result_dict
|
||||
|
||||
def evaluate_scoliosis(data, dataset, metric='euc'):
|
||||
|
||||
msg_mgr = get_msg_mgr()
|
||||
|
||||
from sklearn.metrics import accuracy_score, f1_score, precision_score, recall_score, confusion_matrix
|
||||
|
||||
feature, label, class_id, view = data['embeddings'], data['labels'], data['types'], data['views']
|
||||
|
||||
label = np.array(label)
|
||||
class_id = np.array(class_id)
|
||||
|
||||
# Update class_id with integer labels based on status
|
||||
class_id_int = np.array([1 if status == 'positive' else 2 if status == 'neutral' else 0 for status in class_id])
|
||||
print('class_id=', class_id_int)
|
||||
|
||||
features = np.array(feature)
|
||||
c_id_int = np.argmax(features.mean(-1), axis=-1)
|
||||
print('predicted_labels', c_id_int)
|
||||
|
||||
# Calculate sensitivity and specificity
|
||||
cm = confusion_matrix(class_id_int, c_id_int, labels=[0, 1, 2])
|
||||
FP = cm.sum(axis=0) - np.diag(cm)
|
||||
FN = cm.sum(axis=1) - np.diag(cm)
|
||||
TP = np.diag(cm)
|
||||
TN = cm.sum() - (FP + FN + TP)
|
||||
|
||||
# Sensitivity, hit rate, recall, or true positive rate
|
||||
TPR = TP / (TP + FN)
|
||||
# Specificity or true negative rate
|
||||
TNR = TN / (TN + FP)
|
||||
accuracy = accuracy_score(class_id_int, c_id_int)
|
||||
|
||||
result_dict = {}
|
||||
result_dict["scalar/test_accuracy/"] = accuracy
|
||||
result_dict["scalar/test_sensitivity/"] = TPR
|
||||
result_dict["scalar/test_specificity/"] = TNR
|
||||
|
||||
# Printing the sensitivity and specificity
|
||||
for i, cls in enumerate(['Positive']):
|
||||
print(f"{cls} Sensitivity (Recall): {TPR[i] * 100:.2f}%")
|
||||
print(f"{cls} Specificity: {TNR[i] * 100:.2f}%")
|
||||
print(f"Accuracy: {accuracy * 100:.2f}%")
|
||||
|
||||
return result_dict
|
||||
logits = np.array(data['embeddings'])
|
||||
labels = data['types']
|
||||
|
||||
# Label mapping: negative->0, neutral->1, positive->2
|
||||
label_map = {'negative': 0, 'neutral': 1, 'positive': 2}
|
||||
true_ids = np.array([label_map[status] for status in labels])
|
||||
|
||||
pred_ids = np.argmax(logits.mean(-1), axis=-1)
|
||||
|
||||
# Calculate evaluation metrics
|
||||
# Total Accuracy: proportion of correctly predicted samples among all samples
|
||||
accuracy = accuracy_score(true_ids, pred_ids)
|
||||
|
||||
# Macro-average Precision: average of precision scores for each class
|
||||
precision = precision_score(true_ids, pred_ids, average='macro', zero_division=0)
|
||||
|
||||
# Macro-average Recall: average of recall scores for each class
|
||||
recall = recall_score(true_ids, pred_ids, average='macro', zero_division=0)
|
||||
|
||||
# Macro-average F1: average of F1 scores for each class
|
||||
f1 = f1_score(true_ids, pred_ids, average='macro', zero_division=0)
|
||||
|
||||
# Confusion matrix (for debugging)
|
||||
# cm = confusion_matrix(true_ids, pred_ids, labels=[0, 1, 2])
|
||||
# class_names = ['Negative', 'Neutral', 'Positive']
|
||||
|
||||
# Print results
|
||||
msg_mgr.log_info(f"Total Accuracy: {accuracy*100:.2f}%")
|
||||
msg_mgr.log_info(f"Macro-avg Precision: {precision*100:.2f}%")
|
||||
msg_mgr.log_info(f"Macro-avg Recall: {recall*100:.2f}%")
|
||||
msg_mgr.log_info(f"Macro-avg F1 Score: {f1*100:.2f}%")
|
||||
|
||||
return {
|
||||
"scalar/test_accuracy/": accuracy,
|
||||
"scalar/test_precision/": precision,
|
||||
"scalar/test_recall/": recall,
|
||||
"scalar/test_f1/": f1
|
||||
}
|
||||
|
||||
def evaluate_FreeGait(data, dataset, metric='euc'):
|
||||
msg_mgr = get_msg_mgr()
|
||||
|
||||
@@ -16,10 +16,10 @@ class ScoNet(BaseModel):
|
||||
self.HPP = HorizontalPoolingPyramid(bin_num=model_cfg['bin_num'])
|
||||
|
||||
def forward(self, inputs):
|
||||
ipts, labs, class_id, _, seqL = inputs
|
||||
ipts, pids, labels, _, seqL = inputs
|
||||
|
||||
class_id_int = np.array([1 if status == 'positive' else 2 if status == 'neutral' else 0 for status in class_id])
|
||||
class_id = torch.tensor(class_id_int).cuda()
|
||||
# Label mapping: negative->0, neutral->1, positive->2
|
||||
label_ids = np.array([{'negative': 0, 'neutral': 1, 'positive': 2}[status] for status in labels])
|
||||
|
||||
sils = ipts[0]
|
||||
if len(sils.size()) == 4:
|
||||
@@ -40,8 +40,8 @@ class ScoNet(BaseModel):
|
||||
embed = embed_1
|
||||
retval = {
|
||||
'training_feat': {
|
||||
'triplet': {'embeddings': embed, 'labels': labs},
|
||||
'softmax': {'logits': logits, 'labels': class_id},
|
||||
'triplet': {'embeddings': embed, 'labels': pids},
|
||||
'softmax': {'logits': logits, 'labels': label_ids},
|
||||
},
|
||||
'visual_summary': {
|
||||
'image/sils': rearrange(sils,'n c s h w -> (n s) c h w')
|
||||
|
||||
Reference in New Issue
Block a user