README.md

OpenCV Zoo

A zoo for models tuned for OpenCV DNN with benchmarks on different platforms.

Guidelines:

• To clone this repo, please install git-lfs, run git lfs install and use git lfs clone https://github.com/opencv/opencv_zoo.
• To run benchmark on your hardware settings, please refer to benchmark/README.

Models & Benchmarks

Hardware Setup:

• CPU x86_64: INTEL CPU i7-5930K @ 3.50GHz, 6 cores, 12 threads.
• CPU ARM: Raspberry 4B, BCM2711B0 @ 1.5GHz (Cortex A-72), 4 cores, 4 threads.
• GPU CUDA: NVIDIA Jetson Nano B01, 128-core Maxwell, Quad-core ARM A57 @ 1.43 GHz.

Important Notes:

• The time data that shown on the following table presents the time elapsed from preprocess (resize is excluded), to a forward pass of a network, and postprocess to get final results.
• The time data that shown on the following table is the median of 10 runs. Different metrics may be applied to some specific models.
• View benchmark/config for more details on benchmarking different models.

Model	Input Size	CPU x86_64 (ms)	CPU ARM (ms)	GPU CUDA (ms)*
YuNet	160x120	1.45	6.22	12.18
DB	640x480	137.38	2780.78	220.52
CRNN	100x32	50.21	234.32	196.15
SFace	112x112	8.65	99.20	24.88
PP-ResNet	224x224	56.05	602.58	98.64
PP-HumanSeg	192x192	19.92	105.32	67.97

*: Batch size is 1.

License

OpenCV Zoo is licensed under the Apache 2.0 license. Please refer to licenses of different models.