-
Notifications
You must be signed in to change notification settings - Fork 100
Commit
This commit does not belong to any branch on this repository, and may belong to a fork outside of the repository.
* update models * upload script
- Loading branch information
Showing
9 changed files
with
147 additions
and
5 deletions.
There are no files selected for viewing
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Loading
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
Loading
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
Loading
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
Binary file added
BIN
+233 KB
docs/detection-models/demo/grounding_dino_swin-b_finetune_16xb2_1x_coco_59.7.jpg
Loading
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
Loading
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
Loading
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| Original file line number | Diff line number | Diff line change |
|---|---|---|
| @@ -0,0 +1,69 @@ | ||
| # Download mmdetection and put this script in the directory. | ||
| # git clone https://github.com/open-mmlab/mmdetection | ||
|
|
||
| import os.path | ||
| from glob import glob | ||
|
|
||
| import yaml | ||
| import mmcv | ||
| from mmdet.apis import inference_detector, init_detector | ||
| from mmdet.visualization import DetLocalVisualizer | ||
|
|
||
| img_path = 'demo/123.jpg' | ||
|
|
||
| img = mmcv.imread(img_path) | ||
|
|
||
| model_config_map = {} | ||
| for meta_path in glob('configs/*/metafile.yml'): | ||
| meta = yaml.safe_load(open(meta_path)) | ||
| if not isinstance(meta, dict): | ||
| continue | ||
| for model in meta['Models']: | ||
| model_config_map[model['Config']] = model | ||
|
|
||
| config_path_list = [ | ||
| 'configs/grounding_dino/grounding_dino_swin-b_finetune_16xb2_1x_coco.py', | ||
| 'configs/dino/dino-5scale_swin-l_8xb2-36e_coco.py', | ||
| 'configs/detr/detr_r50_8xb2-150e_coco.py', | ||
| 'configs/faster_rcnn/faster-rcnn_r50_fpn_2x_coco.py', | ||
| 'configs/mask_rcnn/mask-rcnn_r50_fpn_2x_coco.py', | ||
| 'configs/yolo/yolov3_d53_8xb8-ms-608-273e_coco.py', | ||
| ] | ||
|
|
||
| visualizer = DetLocalVisualizer( | ||
| bbox_color=(72, 241, 72), | ||
| text_color=(200, 200, 200), | ||
| mask_color='coco', | ||
| ) | ||
|
|
||
| for config_path in config_path_list: | ||
| model_meta = model_config_map[config_path] | ||
| print(model_meta) | ||
|
|
||
| checkpoint_path = model_meta['Weights'] | ||
| model_filename = os.path.splitext(os.path.split(checkpoint_path)[1])[0] | ||
|
|
||
| score = model_meta['Results'][0]['Metrics']['box AP'] | ||
| model_name = f"{model_meta['Name']}_{score}" | ||
|
|
||
| model = init_detector(config_path, checkpoint_path) | ||
| result = inference_detector(model, img, text_prompt='car') | ||
|
|
||
| visualizer.dataset_meta = model.dataset_meta | ||
| visualizer.add_datasample( | ||
| 'new_result', | ||
| mmcv.bgr2rgb(img), | ||
| data_sample=result, | ||
| draw_gt=False, | ||
| out_file=f'demo/{model_name}.jpg', | ||
| pred_score_thr=0.5) | ||
|
|
||
| # model.show_result( | ||
| # img_path, | ||
| # result, | ||
| # score_thr=0.5, | ||
| # bbox_color=(72, 241, 72), | ||
| # mask_color='coco', | ||
| # text_color=(200, 200, 200), | ||
| # out_file=f'demo/{model_name}.jpg' | ||
| # ) |
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| Original file line number | Diff line number | Diff line change |
|---|---|---|
| @@ -0,0 +1,69 @@ | ||
| # Download mmdetection and put this script in the directory. | ||
| # git clone https://github.com/open-mmlab/mmdetection | ||
|
|
||
| import os.path | ||
| from glob import glob | ||
|
|
||
| import yaml | ||
| import mmcv | ||
| from mmdet.apis import inference_detector, init_detector | ||
| from mmdet.visualization import DetLocalVisualizer | ||
|
|
||
| img_path = 'demo/123.jpg' | ||
|
|
||
| img = mmcv.imread(img_path) | ||
|
|
||
| model_config_map = {} | ||
| for meta_path in glob('configs/*/metafile.yml'): | ||
| meta = yaml.safe_load(open(meta_path)) | ||
| if not isinstance(meta, dict): | ||
| continue | ||
| for model in meta['Models']: | ||
| model_config_map[model['Config']] = model | ||
|
|
||
| config_path_list = [ | ||
| 'configs/grounding_dino/grounding_dino_swin-b_finetune_16xb2_1x_coco.py', | ||
| 'configs/dino/dino-5scale_swin-l_8xb2-36e_coco.py', | ||
| 'configs/detr/detr_r50_8xb2-150e_coco.py', | ||
| 'configs/faster_rcnn/faster-rcnn_r50_fpn_2x_coco.py', | ||
| 'configs/mask_rcnn/mask-rcnn_r50_fpn_2x_coco.py', | ||
| 'configs/yolo/yolov3_d53_8xb8-ms-608-273e_coco.py', | ||
| ] | ||
|
|
||
| visualizer = DetLocalVisualizer( | ||
| bbox_color=(72, 241, 72), | ||
| text_color=(200, 200, 200), | ||
| mask_color='coco', | ||
| ) | ||
|
|
||
| for config_path in config_path_list: | ||
| model_meta = model_config_map[config_path] | ||
| print(model_meta) | ||
|
|
||
| checkpoint_path = model_meta['Weights'] | ||
| model_filename = os.path.splitext(os.path.split(checkpoint_path)[1])[0] | ||
|
|
||
| score = model_meta['Results'][0]['Metrics']['box AP'] | ||
| model_name = f"{model_meta['Name']}_{score}" | ||
|
|
||
| model = init_detector(config_path, checkpoint_path) | ||
| result = inference_detector(model, img, text_prompt='car') | ||
|
|
||
| visualizer.dataset_meta = model.dataset_meta | ||
| visualizer.add_datasample( | ||
| 'new_result', | ||
| mmcv.bgr2rgb(img), | ||
| data_sample=result, | ||
| draw_gt=False, | ||
| out_file=f'demo/{model_name}.jpg', | ||
| pred_score_thr=0.5) | ||
|
|
||
| # model.show_result( | ||
| # img_path, | ||
| # result, | ||
| # score_thr=0.5, | ||
| # bbox_color=(72, 241, 72), | ||
| # mask_color='coco', | ||
| # text_color=(200, 200, 200), | ||
| # out_file=f'demo/{model_name}.jpg' | ||
| # ) |