murrdpirate
murrdpirate
I'm embedding my labels as subtitles using the .srt format. Any thoughts on adding the ability to extract subtitles along with the decoded frames?
The fragment volume is defined as the cubic volume that encloses the view frustums (with each frustum having a max distance). Thus, this volume can change based on the specific...
In my problem, the two point clouds are transformed by only x,y,z position and yaw angle. Would it make sense to modify FGR to enforce this constraint? Should better accuracy/speed...
I have the ability to provide accurate camera pose information, but I'd still like to use DSO's excellent point tracking and depth estimation (in addition to allowing the poses to...
I just downloaded the sdk0.3.x zip and am trying to build the solution. At first, I got errors for all of the .cpp/.h files in src because the EOL format...
This approach seems like it would also be valid for MVS, where there is no fixed baseline between camera views. Has anyone tried that? Any potential concerns with such an...
The torchsparse papers seem focused on outdoor lidar datasets, which are very sparse. I don't see any benchmarks on slightly denser datasets like 3DMatch, where the point clouds are indoor...