Lightweight human detection models generated on high-quality human data sets. It can detect objects with high accuracy and speed in a total of 4 classes: eye, nose, mouth, ear. The resistance to Motion Blur, Gaussian noise, contrast noise, backlighting, and halation is quite strong because it was trained only on images with added photometric noise for all images in the MS-COCO subset of the image set. In addition, about half of the image set was annotated by me with the aspect ratio of the original image substantially destroyed. I manually annotated all images in the dataset by myself. The model is intended to use real-world video for inference and has enhanced resistance to all kinds of noise. Probably stronger than any known model. However, the quality of the known data set and my data set are so different that an accurate comparison of accuracy is not possible.
Don't be ruled by the curse of mAP.
-
Sample
output_eye_nose_mouth_ear_x.mp4
output_eye_nose_mouth_ear_x2.mp4
The use of CD-COCO: Complex Distorted COCO database for Scene-Context-Aware computer vision has also greatly improved resistance to various types of noise.
- Global distortions
- Noise
- Contrast
- Compression
- Photorealistic Rain
- Photorealistic Haze
- Motion-Blur
- Defocus-Blur
- Backlight illumination
- Local distortions
- Motion-Blur
- Defocus-Blur
- Backlight illumination
- COCO-Hand http://vision.cs.stonybrook.edu/~supreeth/COCO-Hand.zip
- CD-COCO: Complex Distorted COCO database for Scene-Context-Aware computer vision
- I am adding my own enhancement data to COCO-Hand and re-annotating all images. In other words, only COCO images were cited and no annotation data were cited.
- I have no plans to publish my own dataset.
- Annotation quantity
TOTAL: 10,883 images TOTAL: 88,769 labels train - 70,378 labels { "eye": 19,379, "nose": 18,385, "mouth": 14,898, "ear": 17,716 } val - 18,391 labels { "eye": 4,999, "nose": 4,925, "mouth": 3,875, "ear": 4,592 }
Halfway compromises are never acceptable.
| Class Name | Class ID |
|---|---|
| Eye | 0 |
| Nose | 1 |
| Mouth | 2 |
| Ear | 3 |
-
Python 3.10
-
onnx 1.14.1+
-
onnxruntime-gpu v1.16.1 (TensorRT Execution Provider Enabled Binary. See: onnxruntime-gpu v1.16.1 + CUDA 11.8 + TensorRT 8.5.3 build (RTX3070))
-
opencv-contrib-python 4.9.0.80
-
numpy 1.24.3
-
TensorRT 8.5.3-1+cuda11.8
# Common ############################################ pip install opencv-contrib-python numpy onnx # For ONNX ########################################## pip uninstall onnxruntime onnxruntime-gpu pip install onnxruntime or pip install onnxruntime-gpu
-
Demonstration of models with built-in post-processing (Float32/Float16)
usage: demo_yolox_onnx_eye_nose_mouth_ear.py \ [-h] \ [-m MODEL] \ (-v VIDEO | -i IMAGES_DIR) \ [-ep {cpu,cuda,tensorrt}] \ [-dvw] \ [-dwk] options: -h, --help show this help message and exit -m MODEL, --model MODEL ONNX/TFLite file path for YOLOX. -v VIDEO, --video VIDEO Video file path or camera index. -i IMAGES_DIR, --images_dir IMAGES_DIR jpg, png images folder path. -ep {cpu,cuda,tensorrt}, \ --execution_provider {cpu,cuda,tensorrt} Execution provider for ONNXRuntime. -dvw, --disable_video_writer Disable video writer. Eliminates the file I/O load associated with automatic recording to MP4. Devices that use a MicroSD card or similar for main storage can speed up overall processing. -dwk, --disable_waitKey Disable cv2.waitKey(). When you want to process a batch of still images, disable key-input wait and process them continuously. -
YOLOX-Wholebody-with-Wheelchair - Nano
Average Precision (AP) @[ IoU=0.50:0.95 | area= all | maxDets=100 ] = 0.212 Average Precision (AP) @[ IoU=0.50 | area= all | maxDets=100 ] = 0.551 Average Precision (AP) @[ IoU=0.75 | area= all | maxDets=100 ] = 0.124 Average Precision (AP) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.197 Average Precision (AP) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.655 Average Precision (AP) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.750 Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets= 1 ] = 0.128 Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets= 10 ] = 0.265 Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets=100 ] = 0.307 Average Recall (AR) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.296 Average Recall (AR) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.698 Average Recall (AR) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.769 per class AP: | class | AP | class | AP | class | AP | |:--------|:-------|:--------|:-------|:--------|:-------| | eye | 16.543 | nose | 24.205 | mouth | 21.591 | | ear | 22.362 | | | | | per class AR: | class | AR | class | AR | class | AR | |:--------|:-------|:--------|:-------|:--------|:-------| | eye | 26.926 | nose | 33.193 | mouth | 30.934 | | ear | 31.600 | | | | | -
YOLOX-Wholebody-with-Wheelchair - Tiny
Average Precision (AP) @[ IoU=0.50:0.95 | area= all | maxDets=100 ] = 0.239 Average Precision (AP) @[ IoU=0.50 | area= all | maxDets=100 ] = 0.609 Average Precision (AP) @[ IoU=0.75 | area= all | maxDets=100 ] = 0.147 Average Precision (AP) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.225 Average Precision (AP) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.685 Average Precision (AP) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.822 Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets= 1 ] = 0.140 Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets= 10 ] = 0.288 Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets=100 ] = 0.325 Average Recall (AR) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.315 Average Recall (AR) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.724 Average Recall (AR) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.833 per class AP: | class | AP | class | AP | class | AP | |:--------|:-------|:--------|:-------|:--------|:-------| | eye | 18.440 | nose | 27.151 | mouth | 24.926 | | ear | 25.282 | | | | | per class AR: | class | AR | class | AR | class | AR | |:--------|:-------|:--------|:-------|:--------|:-------| | eye | 27.932 | nose | 35.623 | mouth | 32.515 | | ear | 34.093 | | | | | -
YOLOX-Wholebody-with-Wheelchair - S
Average Precision (AP) @[ IoU=0.50:0.95 | area= all | maxDets=100 ] = 0.296 Average Precision (AP) @[ IoU=0.50 | area= all | maxDets=100 ] = 0.698 Average Precision (AP) @[ IoU=0.75 | area= all | maxDets=100 ] = 0.202 Average Precision (AP) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.280 Average Precision (AP) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.752 Average Precision (AP) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.877 Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets= 1 ] = 0.161 Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets= 10 ] = 0.339 Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets=100 ] = 0.378 Average Recall (AR) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.368 Average Recall (AR) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.783 Average Recall (AR) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.888 per class AP: | class | AP | class | AP | class | AP | |:--------|:-------|:--------|:-------|:--------|:-------| | eye | 24.173 | nose | 32.757 | mouth | 30.493 | | ear | 30.877 | | | | | per class AR: | class | AR | class | AR | class | AR | |:--------|:-------|:--------|:-------|:--------|:-------| | eye | 32.992 | nose | 40.918 | mouth | 37.490 | | ear | 39.815 | | | | | -
YOLOX-Wholebody-with-Wheelchair - M
Average Precision (AP) @[ IoU=0.50:0.95 | area= all | maxDets=100 ] = 0.322 Average Precision (AP) @[ IoU=0.50 | area= all | maxDets=100 ] = 0.731 Average Precision (AP) @[ IoU=0.75 | area= all | maxDets=100 ] = 0.235 Average Precision (AP) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.307 Average Precision (AP) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.778 Average Precision (AP) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.908 Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets= 1 ] = 0.174 Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets= 10 ] = 0.363 Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets=100 ] = 0.396 Average Recall (AR) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.386 Average Recall (AR) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.806 Average Recall (AR) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.916 per class AP: | class | AP | class | AP | class | AP | |:--------|:-------|:--------|:-------|:--------|:-------| | eye | 26.831 | nose | 35.408 | mouth | 33.207 | | ear | 33.463 | | | | | per class AR: | class | AR | class | AR | class | AR | |:--------|:-------|:--------|:-------|:--------|:-------| | eye | 35.060 | nose | 42.854 | mouth | 39.603 | | ear | 41.065 | | | | | -
YOLOX-Wholebody-with-Wheelchair - L
Average Precision (AP) @[ IoU=0.50:0.95 | area= all | maxDets=100 ] = 0.342 Average Precision (AP) @[ IoU=0.50 | area= all | maxDets=100 ] = 0.758 Average Precision (AP) @[ IoU=0.75 | area= all | maxDets=100 ] = 0.260 Average Precision (AP) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.326 Average Precision (AP) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.777 Average Precision (AP) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.897 Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets= 1 ] = 0.180 Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets= 10 ] = 0.379 Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets=100 ] = 0.412 Average Recall (AR) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.402 Average Recall (AR) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.807 Average Recall (AR) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.911 per class AP: | class | AP | class | AP | class | AP | |:--------|:-------|:--------|:-------|:--------|:-------| | eye | 28.818 | nose | 37.688 | mouth | 34.504 | | ear | 35.616 | | | | | per class AR: | class | AR | class | AR | class | AR | |:--------|:-------|:--------|:-------|:--------|:-------| | eye | 36.814 | nose | 44.787 | mouth | 40.686 | | ear | 42.515 | | | | | -
YOLOX-Wholebody-with-Wheelchair - X
Average Precision (AP) @[ IoU=0.50:0.95 | area= all | maxDets=100 ] = 0.353 Average Precision (AP) @[ IoU=0.50 | area= all | maxDets=100 ] = 0.766 Average Precision (AP) @[ IoU=0.75 | area= all | maxDets=100 ] = 0.274 Average Precision (AP) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.338 Average Precision (AP) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.783 Average Precision (AP) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.919 Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets= 1 ] = 0.186 Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets= 10 ] = 0.389 Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets=100 ] = 0.421 Average Recall (AR) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.410 Average Recall (AR) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.812 Average Recall (AR) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.932 per class AP: | class | AP | class | AP | class | AP | |:--------|:-------|:--------|:-------|:--------|:-------| | eye | 30.084 | nose | 38.311 | mouth | 35.977 | | ear | 36.761 | | | | | per class AR: | class | AR | class | AR | class | AR | |:--------|:-------|:--------|:-------|:--------|:-------| | eye | 37.902 | nose | 45.039 | mouth | 42.007 | | ear | 43.292 | | | | | -
Post-Process
Because I add my own post-processing to the end of the model, which can be inferred by TensorRT, CUDA, and CPU, the benchmarked inference speed is the end-to-end processing speed including all pre-processing and post-processing. EfficientNMS in TensorRT is very slow and should be offloaded to the CPU.
-
NMS default parameter
param value note max_output_boxes_per_class 20 Maximum number of outputs per class of one type. 20indicates that the maximum number of people detected is20, the maximum number of heads detected is20, and the maximum number of hands detected is20. The larger the number, the more people can be detected, but the inference speed slows down slightly due to the larger overhead of NMS processing by the CPU. In addition, as the number of elements in the final output tensor increases, the amount of information transferred between hardware increases, resulting in higher transfer costs on the hardware circuit. Therefore, it would be desirable to set the numerical size to the minimum necessary.iou_threshold 0.40 A value indicating the percentage of occlusion allowed for multiple bounding boxes of the same class. 0.40is excluded from the detection results if, for example, two bounding boxes overlap in more than 41% of the area. The smaller the value, the more occlusion is tolerated, but over-detection may increase.score_threshold 0.25 Bounding box confidence threshold. Specify in the range of 0.00to1.00. The larger the value, the stricter the filtering and the lower the NMS processing load, but in exchange, all but bounding boxes with high confidence values are excluded from detection. This is a parameter that has a very large percentage impact on NMS overhead. -
Change NMS parameters
Use PINTO0309/sam4onnx to rewrite the
NonMaxSuppressionparameter in the ONNX file.For example,
pip install onnxsim==0.4.33 \ && pip install -U simple-onnx-processing-tools \ && pip install -U onnx \ && python -m pip install -U onnx_graphsurgeon \ --index-url https://pypi.ngc.nvidia.com ### max_output_boxes_per_class ### Example of changing the maximum number of detections per class to 100. sam4onnx \ --op_name main01_nonmaxsuppression11 \ --input_onnx_file_path yolox_s_body_head_hand_post_0299_0.4983_1x3x256x320.onnx \ --output_onnx_file_path yolox_s_body_head_hand_post_0299_0.4983_1x3x256x320.onnx \ --input_constants main01_max_output_boxes_per_class int64 [100] ### iou_threshold ### Example of changing the allowable area of occlusion to 20%. sam4onnx \ --op_name main01_nonmaxsuppression11 \ --input_onnx_file_path yolox_s_body_head_hand_post_0299_0.4983_1x3x256x320.onnx \ --output_onnx_file_path yolox_s_body_head_hand_post_0299_0.4983_1x3x256x320.onnx \ --input_constants main01_iou_threshold float32 [0.20] ### score_threshold ### Example of changing the bounding box score threshold to 15%. sam4onnx \ --op_name main01_nonmaxsuppression11 \ --input_onnx_file_path yolox_s_body_head_hand_post_0299_0.4983_1x3x256x320.onnx \ --output_onnx_file_path yolox_s_body_head_hand_post_0299_0.4983_1x3x256x320.onnx \ --input_constants main01_score_threshold float32 [0.15]
-
Post-processing structure
PyTorch alone cannot generate this post-processing.
-
-
INT8 quantization (TexasInstruments/YOLOX-ti-lite)
In my experience, YOLOX has a very large accuracy degradation during quantization due to its structure. The reasons for this and the workaround are examined in detail by TexasInstruments. I have summarized the main points below on how to minimize accuracy degradation during quantization through my own practice. I just put into practice what TexasInstruments suggested, but the degrade in accuracy during quantization was extremely small. Note, however, that the results of the Float16 mixed-precision training before quantization are significantly degraded in accuracy due to the change in activation function to
ReLUand many other workarounds, as well as the completely different data sets being benchmarked.
If this work has contributed in any way to your research or business, I would be happy to be cited in your literature.
@software{YOLOX-Eye-Nose-Mouth-Ear,
author={Katsuya Hyodo},
title={Lightweight human detection models generated on high-quality human data sets. It can detect objects with high accuracy and speed in a total of four classes: eye, nose, mouth, ear.},
url={https://github.com/PINTO0309/PINTO_model_zoo/tree/main/448_YOLOX-Eye-Nose-Mouth-Ear},
year={2024},
month={5},
doi={10.5281/zenodo.10229410}
}I am very grateful for their excellent work.
-
COCO-Hand
https://vision.cs.stonybrook.edu/~supreeth/
@article{Hand-CNN, title={Contextual Attention for Hand Detection in the Wild}, author={Supreeth Narasimhaswamy and Zhengwei Wei and Yang Wang and Justin Zhang and Minh Hoai}, booktitle={International Conference on Computer Vision (ICCV)}, year={2019}, url={https://arxiv.org/pdf/1904.04882.pdf} }
-
CD-COCO: Complex Distorted COCO database for Scene-Context-Aware computer vision
@INPROCEEDINGS{10323035, author={Beghdadi, Ayman and Beghdadi, Azeddine and Mallem, Malik and Beji, Lotfi and Cheikh, Faouzi Alaya}, booktitle={2023 11th European Workshop on Visual Information Processing (EUVIP)}, title={CD-COCO: A Versatile Complex Distorted COCO Database for Scene-Context-Aware Computer Vision}, year={2023}, volume={}, number={}, pages={1-6}, doi={10.1109/EUVIP58404.2023.10323035} }
-
YOLOX
https://github.com/Megvii-BaseDetection/YOLOX
@article{yolox2021, title={YOLOX: Exceeding YOLO Series in 2021}, author={Ge, Zheng and Liu, Songtao and Wang, Feng and Li, Zeming and Sun, Jian}, journal={arXiv preprint arXiv:2107.08430}, year={2021} }
-
YOLOX-ti-lite
-
yolox-ti-lite_tflite
-
YOLOX-Colaboratory-Training-Sample
高橋かずひと https://github.com/Kazuhito00
https://github.com/Kazuhito00/YOLOX-Colaboratory-Training-Sample