CVCUDA / CV-CUDA

CV-CUDA™ is an open-source, GPU accelerated library for cloud-scale image processing and computer vision.
https://cvcuda.github.io
Other
2.39k stars 216 forks source link

[QUESTION] Error in cvtColor,Invalid input channel number #174

Closed csrddbb closed 2 months ago

csrddbb commented 5 months ago

//: # ""

//: # ""

What is your question? ` nvcv::Image image(nvcv::Size2D(w, h), nvcv::FMT_YUYV); nvcv::ImageBatchVarShape image_batch(1); image_batch.pushBack(image);

nvcv::Image cvt_image(nvcv::Size2D(w, h), nvcv::FMT_BGR8); nvcv::ImageBatchVarShape cvt_image_batch(1); cvt_image_batch.pushBack(cvt_image);

cvcuda::CvtColor cvt_op; cvt_op(stream, image_batch, cvt_image_batch, NVCV_COLOR_YUV2BGR_YUYV); `

Why does converting YUYV format to BGR result in a channel number error? YUYV format has 3 channels. What is the reason? How should I modify the code?

output error: ERROR: Invalid input channel number 3

enmortensen commented 2 months ago

Hi @csrddbb,

Thank you for your interest is CV-CUDA.

This is a known bug that will be fixed in the next release, v0.12.0, at the end of September.