Bin-ze
Bin-ze
I have the same problem, have you found a solution?
这个bug为什么转换yolov5不出现,还有使用pip直接安装tensorrt不出现,只出现在自己源码安装tensorrt或许jston平台刷的tensorrt也有问题,感觉很奇怪 发自我的iPhone > 在 2022年8月25日,01:10,Monday-Leo ***@***.***> 写道: > > > 因为tensorrt 8.2有bug。升级至8.4可以解决。 > > — > Reply to this email directly, view it on GitHub, or unsubscribe. > You...
那我自己写nms,倒出无nms的onnx模型呢 发自我的iPhone > 在 2022年8月25日,18:35,Monday-Leo ***@***.***> 写道: > > > 这是nms的bug,yolov5的nms没有并入到模型当中,所以没关系。 > > — > Reply to this email directly, view it on GitHub, or unsubscribe. > You are...
tesla T4 发自我的iPhone > 在 2022年9月16日,14:00,stoooner ***@***.***> 写道: > > @Bin-ze what is your GPU? > > — > Reply to this email directly, view it on GitHub, or...
In fact, I test the speed of a single picture and remove the post-processing can reach more than 100fps, but when I add nms and pre-processing, the speed is less...
不是英语的问题,是代码写的确实很烂,readme交代的也会非常不清楚,自己的数据集写的dataset类,简直了,写的很抽象。
可以参考这里:https://github.com/Bin-ze/BEVFormer_segmentation_detection
> 您可以尝试以下代码,它在模型的捆绑调整和重新三角测量/轨道合并之间交替。运行几次迭代应该有助于完善不太准确的姿势: > > ``` > int RunIncrementalModelRefiner(int argc, char** argv) { > std::string input_path; > std::string output_path; > std::string image_list_path; > > OptionManager options; > options.AddDatabaseOptions(); > options.AddImageOptions(); >...
@ingra14m Thank you very much for your implementation of depth forward and direction, but when I pull your code and then try to do depth supervision, a memory leak occurs...
> There are indeed many artifacts in the depth maps of 3D-GS. I tried to improve the geometry of 3D-GS by adding depth loss in [#5](https://github.com/graphdeco-inria/diff-gaussian-rasterization/pull/5) in [diff-gaussian-rasterization](https://github.com/graphdeco-inria/diff-gaussian-rasterization/), which might...