Dinh-Cuong Hoang

@francescooliva Thank you for sharing your experience. I'll try it.

@Birdylx Thank you!. Other than that, can I increase the input size? The current size is: 300x300.

@Birdylx Thank you for the quick reply. I modified the file mobilenetv1_ssd_config.py to: image_size = 512 image_mean = np.array([127, 127, 127]) # RGB layout image_std = 128.0 iou_threshold = 0.45...

@Birdylx The same, getting error when change the image size to 512 as mentioned in the paper

@Birdylx Thank you for the suggestion. It works! However I still fails to detect small objects and big objects at the same time. I modified the SSDBoxSizes with smaller number...

Hi, It should be faster if you put multiple images into a batch and operate the batch. In addition, there is number of things to consider if you want to...

Hi, Thanks for your reply. I Played around and tried to have a better registration result. As you can see in the image below, it looks good. But, the label...

Hi, Not yet. But, everything I need working fine without solving this problem.

openni2-camera-lcm doesn't work for me to record log file. Solution: install rgbd_ros_to_lcm (code from https://github.com/MobileManipulation/rgbd_ros_to_lcm), Then, to record log file: Run launch file in catkin $ roslaunch rgbd_ros_to_lcm lcm_republisher.launch In...

I met some problems related to cmake file, and made some modification and it worked. https://github.com/hoangcuongbk80/Generate-LCM-LabelFusion