HuanyuanZhou
commented
3 years ago Hi, I meet the same problem. Could you tell me how you solve this problem?
Closed morgenzhang closed 5 years ago
HuanyuanZhou
commented
3 years ago Hi, I meet the same problem. Could you tell me how you solve this problem?
Hi, I want to apply InfiniTAM to do 3D mapping on KITTI odometry dataset. I have pre-computed the depth map of each frame and also want to use the ground truth poses as input. I find that you have already provided such tracker named ITMFileBasedTracker. So I changed the trackerConfig in ITMLibSetting to: trackerConfig = "type=file, mask=/path/to/gt_pose/sequencenum/%04i.txt"; I also added some script to see if each pose has been readed correctly:
processing one frame (0)... Loading odometry ground truth from file: /home/zhangyic/Downloads/Semantic_Mapping/kitti_data/gt_pose/01/0000.txt Current pose in file based tracker: 1, 0, 0, 0 0, 1, 0, 0 0, 0, 1, 0 0, 0, 0, 1
processing one frame (1)... Loading odometry ground truth from file: /home/zhangyic/Downloads/Semantic_Mapping/kitti_data/gt_pose/01/0001.txt Current pose in file based tracker: 0.99905, -0.00165, 0.043552, 0.051547 0.00176, 0.999995, -0.002502, -0.024249 -0.043548, 0.002577, 0.999048, 1.00073 0, 0, 0, 1 ... However, when I started the InfiniTAM, and selected by default "follow camera", the output view can not display any result (as shown below)
If I typed f and change to free view mode, I can somehow see the result (as shown below)
Can you explain why this happened? Or whether I have made some mistakes when using this filebasedtracker? If so, I hope you can give me some instruction on how to use it. Thank you very much for your time.