Chen Feng

Found the same error while using matplotlib through MobaXterm. Using the following answer (turning off the RANDR in mobaxterm's x11 configuration) fixed the issue: https://stackoverflow.com/questions/51218417/qxcbconnection-xcb-error-145-unknown-error-message-starting-spyder-with-moba

@ducha-aiki In the SELU paper they seem to suggest using MSRA-like initialization for weights, while your prototxt seem to use fixed Gaussian initialization? Could this affect the result?

Got it, so I guess selu might be only good for FNN instead of CNN? ...... 2017年8月21日 02:16，"Dmytro Mishkin" 写道： > @simbaforrest I don't use fixed > initialization, I use...

@untom Anyone know whether cuSOLVER's Cholesky factorization still suffers for the memory leak in CUDA 7.5?

I think this doesn't solve the problem. If I want to render a mesh with face_colors=[0,200,0], what I get is [0,228,0] (same for vertex_colors). @mmatl it would be great to...

Oh, forget about my previous comment. Just notice this post: https://github.com/mmatl/pyrender/issues/15#issuecomment-480400477 Thanks again for this great library.

> @simbaforrest do you have a sample of code replicating the problem you see (with [0,200,0] becoming [0,228,0])? The comment in #15 is a solution only in the context of...

Thanks, @Cogito2012. We will discuss with the team and get back to you this next week after the ICRA deadline. @RoboticsYimingLi

You can check the deeplab implementation which has a ResNet101 for segmentation.

@wuyouqian3 and others, if @dmax123's answer addressed your questions, please consider closing this issue. Meanwhile, I would suggest you take a look at our newest CVPR2023 work along this line...