Object Detection - Quick Start¶
Object detection is the process of identifying and localizing objects in an image and is an important task in computer vision. Follow this tutorial to learn how to use AutoGluon for object detection.
Tip: If you are new to AutoGluon, review Image Prediction - Quick Start first to learn the basics of the AutoGluon API.
Our goal is to detect motorbike in images by YOLOv3 model. A tiny dataset is collected from VOC dataset, which only contains the motorbike category. The model pretrained on the COCO dataset is used to fine-tune our small dataset. With the help of AutoGluon, we are able to try many models with different hyperparameters automatically, and return the best one as our final model.
To start, import ObjectDetector:
from autogluon.vision import ObjectDetector
/var/lib/jenkins/miniconda3/envs/autogluon-tutorial-object-detection-v3/lib/python3.9/site-packages/gluoncv/__init__.py:40: UserWarning: Both mxnet==1.7.0 and torch==1.10.1+cu102 are installed. You might encounter increased GPU memory footprint if both framework are used at the same time.
warnings.warn(f'Both mxnet=={mx.__version__} and torch=={torch.__version__} are installed. '
Tiny_motorbike Dataset¶
We collect a toy dataset for detecting motorbikes in images. From the VOC dataset, images are randomly selected for training, validation, and testing - 120 images for training, 50 images for validation, and 50 for testing. This tiny dataset follows the same format as VOC.
Using the commands below, we can download this dataset, which is only
23M. The name of unzipped folder is called tiny_motorbike. Anyway,
the task dataset helper can perform the download and extraction
automatically, and load the dataset according to the detection formats.
url = 'https://autogluon.s3.amazonaws.com/datasets/tiny_motorbike.zip'
dataset_train = ObjectDetector.Dataset.from_voc(url, splits='trainval')
tiny_motorbike/
├── Annotations/
├── ImageSets/
└── JPEGImages/
Fit Models by AutoGluon¶
In this section, we demonstrate how to apply AutoGluon to fit our detection models. We use mobilenet as the backbone for the YOLOv3 model. Two different learning rates are used to fine-tune the network. The best model is the one that obtains the best performance on the validation dataset. You can also try using more networks and hyperparameters to create a larger searching space.
We fit a classifier using AutoGluon as follows. In each experiment
(one trial in our searching space), we train the model for 5 epochs to
avoid bursting our tutorial runtime.
time_limit = 60*30 # at most 0.5 hour
detector = ObjectDetector()
hyperparameters = {'epochs': 5, 'batch_size': 8}
hyperparameter_tune_kwargs={'num_trials': 2}
detector.fit(dataset_train, time_limit=time_limit, hyperparameters=hyperparameters, hyperparameter_tune_kwargs=hyperparameter_tune_kwargs)
The number of requested GPUs is greater than the number of available GPUs.Reduce the number to 1
Randomly split train_data into train[158]/validation[12] splits.
Starting HPO experiments
0%| | 0/2 [00:00<?, ?it/s]
modified configs(<old> != <new>): {
root.gpus (0, 1, 2, 3) != (0,)
root.valid.batch_size 16 != 8
root.ssd.base_network vgg16_atrous != resnet50_v1
root.ssd.data_shape 300 != 512
root.dataset_root ~/.mxnet/datasets/ != auto
root.num_workers 4 != 8
root.train.epochs 20 != 5
root.train.seed 233 != 588
root.train.early_stop_patience -1 != 10
root.train.early_stop_max_value 1.0 != inf
root.train.batch_size 16 != 8
root.train.early_stop_baseline 0.0 != -inf
root.dataset voc_tiny != auto
}
Saved config to /var/lib/jenkins/workspace/workspace/autogluon-tutorial-object-detection-v3/docs/_build/eval/tutorials/object_detection/cea9627b/.trial_0/config.yaml
Using transfer learning from ssd_512_resnet50_v1_coco, the other network parameters are ignored.
Start training from [Epoch 0]
[Epoch 0] Training cost: 9.329026, CrossEntropy=3.629686, SmoothL1=1.035936
[Epoch 0] Validation:
bus=nan
chair=0.0
pottedplant=nan
person=0.6231751007613078
cow=nan
dog=nan
bicycle=0.0
boat=nan
motorbike=0.4718204467312355
car=0.8831168831168833
mAP=0.39562248612188533
[Epoch 0] Current best map: 0.395622 vs previous 0.000000, saved to /var/lib/jenkins/workspace/workspace/autogluon-tutorial-object-detection-v3/docs/_build/eval/tutorials/object_detection/cea9627b/.trial_0/best_checkpoint.pkl
[Epoch 1] Training cost: 8.257013, CrossEntropy=2.783386, SmoothL1=1.261059
[Epoch 1] Validation:
bus=nan
chair=0.0
pottedplant=nan
person=0.6296870217324763
cow=nan
dog=nan
bicycle=0.0
boat=nan
motorbike=0.518920916481892
car=0.7537190082644629
mAP=0.38046538929576623
[Epoch 2] Training cost: 8.192107, CrossEntropy=2.691025, SmoothL1=1.179705
[Epoch 2] Validation:
bus=nan
chair=0.0
pottedplant=nan
person=0.6974957912457911
cow=nan
dog=nan
bicycle=0.0
boat=nan
motorbike=0.65718684167882
car=0.34090909090909094
mAP=0.3391183447667404
[Epoch 3] Training cost: 8.580774, CrossEntropy=2.307059, SmoothL1=1.003880
[Epoch 3] Validation:
bus=nan
chair=0.0
pottedplant=nan
person=0.7572150072150072
cow=nan
dog=nan
bicycle=0.0
boat=nan
motorbike=0.7360773634967183
car=0.7459893048128342
mAP=0.44785633510491196
[Epoch 3] Current best map: 0.447856 vs previous 0.395622, saved to /var/lib/jenkins/workspace/workspace/autogluon-tutorial-object-detection-v3/docs/_build/eval/tutorials/object_detection/cea9627b/.trial_0/best_checkpoint.pkl
[Epoch 4] Training cost: 8.084239, CrossEntropy=2.088778, SmoothL1=0.958997
[Epoch 4] Validation:
bus=nan
chair=0.0
pottedplant=nan
person=0.7526355996944232
cow=nan
dog=nan
bicycle=0.0
boat=nan
motorbike=0.6535268957517761
car=0.7142857142857143
mAP=0.42408964194638277
Applying the state from the best checkpoint...
Finished, total runtime is 66.20 s
{ 'best_config': { 'dataset': 'auto',
'dataset_root': 'auto',
'estimator': <class 'gluoncv.auto.estimators.ssd.ssd.SSDEstimator'>,
'gpus': [0],
'horovod': False,
'num_workers': 8,
'resume': '',
'save_interval': 1,
'ssd': { 'amp': False,
'base_network': 'resnet50_v1',
'data_shape': 512,
'filters': None,
'nms_thresh': 0.45,
'nms_topk': 400,
'ratios': ( [1, 2, 0.5],
[1, 2, 0.5, 3, 0.3333333333333333],
[1, 2, 0.5, 3, 0.3333333333333333],
[1, 2, 0.5, 3, 0.3333333333333333],
[1, 2, 0.5],
[1, 2, 0.5]),
'sizes': (30, 60, 111, 162, 213, 264, 315),
'steps': (8, 16, 32, 64, 100, 300),
'syncbn': False,
'transfer': 'ssd_512_resnet50_v1_coco'},
'train': { 'batch_size': 8,
'dali': False,
'early_stop_baseline': -inf,
'early_stop_max_value': inf,
'early_stop_min_delta': 0.001,
'early_stop_patience': 10,
'epochs': 5,
'log_interval': 100,
'lr': 0.001,
'lr_decay': 0.1,
'lr_decay_epoch': (160, 200),
'momentum': 0.9,
'seed': 588,
'start_epoch': 0,
'wd': 0.0005},
'valid': { 'batch_size': 8,
'iou_thresh': 0.5,
'metric': 'voc07',
'val_interval': 1}},
'total_time': 66.1982650756836,
'train_map': 0.7848649650103364,
'valid_map': 0.44785633510491196}
<autogluon.vision.detector.detector.ObjectDetector at 0x7fa113e97c70>
Note that num_trials=2 above is only used to speed up the tutorial.
In normal practice, it is common to only use time_limit and drop
num_trials. Also note that hyperparameter tuning defaults to random
search.
After fitting, AutoGluon automatically returns the best model among all models in the searching space. From the output, we know the best model is the one trained with the second learning rate. To see how well the returned model performed on test dataset, call detector.evaluate().
dataset_test = ObjectDetector.Dataset.from_voc(url, splits='test')
test_map = detector.evaluate(dataset_test)
print("mAP on test dataset: {}".format(test_map[1][-1]))
tiny_motorbike/
├── Annotations/
├── ImageSets/
└── JPEGImages/
mAP on test dataset: 0.024272081734926815
Below, we randomly select an image from test dataset and show the
predicted class, box and probability over the origin image, stored in
predict_class, predict_rois and predict_score columns,
respectively. You can interpret predict_rois as a dict of (xmin,
ymin, xmax, ymax) proportional to original image size.
image_path = dataset_test.iloc[0]['image']
result = detector.predict(image_path)
print(result)
predict_class predict_score 0 person 0.994859
1 motorbike 0.897595
2 bicycle 0.159514
3 car 0.108762
4 motorbike 0.107096
.. ... ...
66 car 0.032677
67 person 0.032589
68 person 0.032493
69 person 0.032453
70 chair 0.032441
predict_rois
0 {'xmin': 0.40365543961524963, 'ymin': 0.278845...
1 {'xmin': 0.31720587611198425, 'ymin': 0.447850...
2 {'xmin': 0.3179326057434082, 'ymin': 0.4476031...
3 {'xmin': 0.0, 'ymin': 0.6284662485122681, 'xma...
4 {'xmin': 0.0, 'ymin': 0.6287426352500916, 'xma...
.. ...
66 {'xmin': 0.3179326057434082, 'ymin': 0.4476031...
67 {'xmin': 0.3488602340221405, 'ymin': 0.2295106...
68 {'xmin': 0.4020317494869232, 'ymin': 0.3283315...
69 {'xmin': 0.3850856423377991, 'ymin': 0.3481267...
70 {'xmin': 0.9034497141838074, 'ymin': 0.0723451...
[71 rows x 3 columns]
Prediction with multiple images is permitted:
bulk_result = detector.predict(dataset_test)
print(bulk_result)
predict_class predict_score 0 person 0.994859
1 motorbike 0.897595
2 bicycle 0.159514
3 car 0.108762
4 motorbike 0.107096
... ... ...
4044 person 0.025370
4045 person 0.025097
4046 person 0.025089
4047 person 0.025078
4048 person 0.024939
predict_rois 0 {'xmin': 0.40365543961524963, 'ymin': 0.278845...
1 {'xmin': 0.31720587611198425, 'ymin': 0.447850...
2 {'xmin': 0.3179326057434082, 'ymin': 0.4476031...
3 {'xmin': 0.0, 'ymin': 0.6284662485122681, 'xma...
4 {'xmin': 0.0, 'ymin': 0.6287426352500916, 'xma...
... ...
4044 {'xmin': 0.3165423274040222, 'ymin': 0.6834895...
4045 {'xmin': 0.4035860300064087, 'ymin': 0.0868127...
4046 {'xmin': 0.2324240803718567, 'ymin': 0.0590429...
4047 {'xmin': 0.3869120478630066, 'ymin': 0.1109670...
4048 {'xmin': 0.2259625345468521, 'ymin': 0.9872213...
image
0 /var/lib/jenkins/.gluoncv/datasets/tiny_motorb...
1 /var/lib/jenkins/.gluoncv/datasets/tiny_motorb...
2 /var/lib/jenkins/.gluoncv/datasets/tiny_motorb...
3 /var/lib/jenkins/.gluoncv/datasets/tiny_motorb...
4 /var/lib/jenkins/.gluoncv/datasets/tiny_motorb...
... ...
4044 /var/lib/jenkins/.gluoncv/datasets/tiny_motorb...
4045 /var/lib/jenkins/.gluoncv/datasets/tiny_motorb...
4046 /var/lib/jenkins/.gluoncv/datasets/tiny_motorb...
4047 /var/lib/jenkins/.gluoncv/datasets/tiny_motorb...
4048 /var/lib/jenkins/.gluoncv/datasets/tiny_motorb...
[4049 rows x 4 columns]
We can also save the trained model, and use it later.
savefile = 'detector.ag'
detector.save(savefile)
new_detector = ObjectDetector.load(savefile)
/var/lib/jenkins/miniconda3/envs/autogluon-tutorial-object-detection-v3/lib/python3.9/site-packages/mxnet/gluon/block.py:1512: UserWarning: Cannot decide type for the following arguments. Consider providing them as input:
data: None
input_sym_arg_type = in_param.infer_type()[0]