OpenDriveLab / ThinkTwice

[CVPR 2023] Pytorch implementation of ThinkTwice, a SOTA Decoder for End-to-end Autonomous Driving under BEV.

OpenDriveLab/ThinkTwice Issues

data collection
Updated 2 months ago
About collect images for calibration.
Closed 3 months ago4
How to set parameters of transform.py?
Closed 4 months ago4
the inference time
Closed 4 months ago1
Performance gap in reproducing results
Updated 5 months ago3
KeyError: 'BENCHMARK'
Closed 5 months ago2
Eval error: Segmentation fault (core dumped)
Closed a year ago4
Question about .register_module() function
Closed 8 months ago4
questions about LIDAR2CAM
Closed 8 months ago4
The versions of the nvidia gpu, driver, cuda, cudnn?
Closed 9 months ago2
speeds of other vehicles
Closed a year ago1
'3d_bbs'
Closed a year ago1
_get_3d_bbs
Closed a year ago1
Error when collecting data
Closed a year ago1
More details on the Longest6 benchmark
Closed a year ago1
data generation
Closed a year ago
Carla failed to start on remote server
Closed a year ago4
ImportError: cannot import name 'voxel_pooling_ext'
Closed a year ago
Question about the bounding box 3d visulization.
Closed a year ago1
Does ThinkTwice have Evaluation on Carla Leaderboard?
Closed a year ago11
Research details
Closed a year ago1
when I compile CUDA function for LSS from BEVDepth，it repears the problem like followings:
Closed a year ago5