This forum is disabled, please visit https://forum.opencv.org
 | 1 | initial version |
@Rbt, Yeah, OpenCV can import models with fp16 weights (Caffe/TensorFlow) or uint8 (TensorFlow) into fp32 format. All the computations are in single precision floats (fp32). There is a PR with FP16 support for Intel's GPUs: https://github.com/opencv/opencv/pull/11062. Read more in OpenCV's evolution proposal: https://github.com/opencv/opencv/issues/11009.
