Vision Model List

List of supported computer vision models.

These are the “well-known” computer vision models that wasmVision can automatically download and run.

candy-8

Candy model for fast neural style transfer (opset 8)

For more information see:

https://github.com/onnx/models/tree/main/validated/vision/style_transfer/fast_neural_style

candy-9

Candy model for fast neural style transfer (opset 8)

For more information see:

https://github.com/onnx/models/tree/main/validated/vision/style_transfer/fast_neural_style

dexined_2024sep

Dense Extreme Inception Network for Edge Detection (DexiNed), a Convolutional Neural Network (CNN) architecture for edge detection.

For more information see:

https://github.com/opencv/opencv_zoo/blob/main/models/edge_detection_dexined/README.md

and also

https://github.com/xavysp/DexiNed

mobilefacenet_2022july

MobileFaceNet model for Facial Expression Recognition (FER) featuring semi-supervised learning model with 88.27% accuracy.

For more information see https://github.com/opencv/opencv_zoo/blob/main/models/facial_expression_recognition/README.md

mobilefacenet_2022july_int8

MobileFaceNet model for Facial Expression Recognition (quantized)

For more information see https://github.com/opencv/opencv_zoo/blob/main/models/facial_expression_recognition/README.md

mobilefacenet_2022july_int8bq

MobileFaceNet model for Facial Expression Recognition (quantized with bias)

For more information see https://github.com/opencv/opencv_zoo/blob/main/models/facial_expression_recognition/README.md

mosaic-8

Mosaic model for fast neural style transfer (opset 8)

For more information see:

https://github.com/onnx/models/tree/main/validated/vision/style_transfer/fast_neural_style

mosaic-9

Mosaic model for fast neural style transfer (opset 9)

For more information see:

https://github.com/onnx/models/tree/main/validated/vision/style_transfer/fast_neural_style

pointilism-8

Pointilism model for fast neural style transfer (opset 8)

For more information see:

https://github.com/onnx/models/tree/main/validated/vision/style_transfer/fast_neural_style

pointilism-9

Pointilism model for fast neural style transfer (opset 9)

For more information see:

https://github.com/onnx/models/tree/main/validated/vision/style_transfer/fast_neural_style

rain-princess-8

Rain Princess model for fast neural style transfer (opset 8)

For more information see:

https://github.com/onnx/models/tree/main/validated/vision/style_transfer/fast_neural_style

rain-princess-9

Rain Princess model for fast neural style transfer (opset 9)

For more information see:

https://github.com/onnx/models/tree/main/validated/vision/style_transfer/fast_neural_style

udnie-8

Udnie model for fast neural style transfer (opset 8)

For more information see:

https://github.com/onnx/models/tree/main/validated/vision/style_transfer/fast_neural_style

udnie-9

Udnie model for fast neural style transfer (opset 9)

For more information see:

https://github.com/onnx/models/tree/main/validated/vision/style_transfer/fast_neural_style

yolov8l

YOLOv8 real-time object detection model. YOLOv8 was released by Ultralytics on January 10th, 2023, offering cutting-edge performance in terms of accuracy and speed.

This is the YOLOv8l variant.

For more information see:

https://docs.ultralytics.com/models/yolov8/

yolov8m

YOLOv8 real-time object detection model. YOLOv8 was released by Ultralytics on January 10th, 2023, offering cutting-edge performance in terms of accuracy and speed.

This is the YOLOv8m variant.

For more information see:

https://docs.ultralytics.com/models/yolov8/

yolov8n

YOLOv8 real-time object detection model. YOLOv8 was released by Ultralytics on January 10th, 2023, offering cutting-edge performance in terms of accuracy and speed.

This is the YOLOv8n variant.

For more information see:

https://docs.ultralytics.com/models/yolov8/

yolov8s

YOLOv8 real-time object detection model. YOLOv8 was released by Ultralytics on January 10th, 2023, offering cutting-edge performance in terms of accuracy and speed.

This is the YOLOv8s variant.

For more information see:

https://docs.ultralytics.com/models/yolov8/

yolov8x

YOLOv8x model for object detection

YOLOv8 real-time object detection model. YOLOv8 was released by Ultralytics on January 10th, 2023, offering cutting-edge performance in terms of accuracy and speed.

This is the YOLOv8s variant.

For more information see:

https://docs.ultralytics.com/models/yolov8/

yolox

YOLOX is an anchor-free version of YOLO, with a simpler design but better performance! It aims to bridge the gap between research and industrial communities. YOLOX is a high-performing object detector, an improvement to the existing YOLO series. YOLO series are in constant exploration of techniques to improve the object detection techniques for optimal speed and accuracy trade-off for real-time applications.

For more information see:

https://github.com/opencv/opencv_zoo/tree/main/models/object_detection_yolox

yunet_2023mar

YuNet, a light-weight, fast and accurate face detection model, which achieves 0.834(AP_easy), 0.824(AP_medium), 0.708(AP_hard) on the WIDER Face validation set.

For more information see:

https://github.com/opencv/opencv_zoo/blob/main/models/face_detection_yunet/README.md