RobustVideoMatting
RobustVideoMatting copied to clipboard
PeterL1n
Robust Video Matting in PyTorch, TensorFlow, TensorFlow.js, ONNX, CoreML!
#159 我遇到了和该问题一致的情况，在使用v100时，模型推理速度只有2-3帧，speedtest却有60-70帧，请问是以下代码造成这种问题吗 ``` if output_foreground is not None: writer_fgr.write(fgr[0]) if output_alpha is not None: writer_pha.write(pha[0]) # pha[0] [12,1,h,w] if output_composition is not None: if output_type == 'video': com = fgr...
`import argparse import torch from torch.utils.mobile_optimizer import optimize_for_mobile from model import MattingNetwork import torch.utils.data.distributed rec = (torch.zeros([1, 1, 1, 1, 1]).to("cuda"),) * 4 src = torch.randn(1, 1, 3, 1080, 1920).to("cuda")...
Thank you for your work! I have a question about license. Your code is licensed under GPLv3. But what about your pre-trained models? Are they licensed under GPLv3, too?
Hi, I want to know why the inference is not running on my GPU, in fact, it is only running on my CPU, are there any requirements or drivers I...
我尝试使用RVM的训练代码进行结果的复现，但是在训练代码的初始阶段加入固定种子的代码段后训练过程中依然出现了偏差。 加入的代码段如下：  请问作者大大是否遇到过类似的问题或有解决这种问题的经验？
Hi，我在Android arm64-v8a平台上，用C++测试两个模型rvm_mobilenetv3_fp16.onnx和rvm_mobilenetv3_fp32.onnx， 发现推理速度差不多，请问这是怎么回事？
Great project, THX something in hand will be detected as background sometimes, so i want to finetune the model，but now， only some images with alpha was find，so if i any...
