# affine transform for feature extraction img, trans, mask, bbox = self.affine_transform_and_masking(img, mask, cfg.render_img_shape, mask, expand_ratio=cfg.render_bbox_expand_ratio) K[0, 0] = K[0, 0] * cfg.render_img_shape[1] / bbox[2] K[1, 1] = K[1, 1] * cfg.render_img_shape[0] / bbox[3] K[0, 2] = (K[0, 2] - bbox[0]) * cfg.render_img_shape[1] / bbox[2] K[1, 2] = (K[1, 2] - bbox[1]) * cfg.rende..
https://github.com/microsoft/torchgeo/issues/529 Conda installs a cpu version of pytorch and torchvision · Issue #529 · microsoft/torchgeo Conda installation (tested on Ubuntu 20.04) installs a cpu version of pytorch and torchvision. In my own project, the only remedy I've found was to hardcode the pytorch package to use. Poor sol... github.com 'conda install pytorch' kept throwed me torch.conda..
Why do they make a new camera coordinate system... so confusing. TL;DR . CO3D uses the Pytorch3D camera coordinate system, and it is different from the OpenGL camera coordinate system and the conventional camera coordinate system. CO3D also uses the "ndc_norm_image_bounds" system for the pixel space. It normalizes the pixel coordinates to [-1,1]. Also, the rotation should be multiplied from the ..
https://logicatcore.github.io/scratchpad/lidar/sensor-fusion/jupyter/2021/04/20/3D-Oriented-Bounding-Box.html 3D Oriented bounding boxes made simple Calculating 3D oriented bounding boxes. Oriented boxes are useful to avoid obstacles and make best utilitsation of the real navigationable space for autonomous vehicles to steer around. logicatcore.github.io 이거 어떻게 하지 고민했었는데, data point 의 eigen vect..
https://github.com/NVIDIA/apex/issues/473
https://github.com/ga0/pyprotect/issues/11 libpyprotect.so: undefined symbol: PyThread_tss_get · Issue #11 · ga0/pyprotect After make, i get the file "libpyprotect.cpython-37m-x86_64-linux-gnu.so", then from the same directory i do "import libpyprotect" i get ModuleNotFound error, Then I rename it t... github.com https://github.com/hfutcgncas/normalSpeed/issues/3
https://pybullet.org/Bullet/phpBB3/viewtopic.php?t=8562
Note: aspect_ratio = tan(0.5*xfov) / tan(0.5*yfov) f_x = (img_width / 2) / tan(xfov / 2) https://pycollada.readthedocs.io/en/latest/reference/generated/collada.camera.PerspectiveCamera.html collada.camera.PerspectiveCamera — pycollada 0.7 documentation Parameters: id (str) – Identifier for the camera znear (float) – Distance to the near clipping plane zfar (float) – Distance to the far clipping ..
- Total
- Today
- Yesterday
- pytorch
- Transformation
- world coordinate
- 에디톨로지
- spin
- 컴퓨터비젼
- Generative model
- densepose
- Virtual Camera
- 인터뷰
- part segmentation
- pyrender
- VAE
- 컴퓨터비전
- nohup
- demo
- 비전
- nerf
- Docker
- deep learning
- 2d pose
- focal length
- Pose2Mesh
- 피트니스
- 문경식
- camera coordinate
- 헬스
- Machine Learning
- Interview
- 머신러닝
일 | 월 | 화 | 수 | 목 | 금 | 토 |
---|---|---|---|---|---|---|
1 | 2 | 3 | 4 | |||
5 | 6 | 7 | 8 | 9 | 10 | 11 |
12 | 13 | 14 | 15 | 16 | 17 | 18 |
19 | 20 | 21 | 22 | 23 | 24 | 25 |
26 | 27 | 28 | 29 | 30 | 31 |