Official PyTorch implementation of DD3D: Is Pseudo-Lidar needed for Monocular 3D Object detection? (ICCV 2021), Dennis Park*, Rares Ambrus*, Vitor Guizilini, Jie Li, and Adrien Gaidon.
Hello @dennis-park-TRI , thank you for your work in DD3D! That's very insightful!
As I go through the repo, I can find some code related to depth pertaining like DenseDepthHead and so on. However, I didn't find such a config to start the depth pretraining. Could you provide your config file for pretraining depth on the V99 model? I assume the data format can be in KITTI since it provides Velodyne data.
Again, thank you for opensource this awesome repo!
I would also be interested in having the config file for depth pre-training. I'm wondering if I can fine-tune the depth pretrained model with additional depth data.
Hello @dennis-park-TRI , thank you for your work in DD3D! That's very insightful!
As I go through the repo, I can find some code related to depth pertaining like DenseDepthHead and so on. However, I didn't find such a config to start the depth pretraining. Could you provide your config file for pretraining depth on the V99 model? I assume the data format can be in KITTI since it provides Velodyne data.
Again, thank you for opensource this awesome repo!