Hi, Wenbo Wen!
Thank you for sharing your code. I have some questions about the production of the dataset, and I hope to get your answers. The first question is, when I want to estimate the pose of my own objects such as gears, is there a tutorial on how to make a dataset of gears to apply to your network? The second question is, when I record a video of a gear moving, how do I transmit the video to the video segmentation network mentioned in your paper to achieve the output of spatio-temporal consistent posture?
Looking forward to your reply, thank you again!
Hi, Wenbo Wen! Thank you for sharing your code. I have some questions about the production of the dataset, and I hope to get your answers. The first question is, when I want to estimate the pose of my own objects such as gears, is there a tutorial on how to make a dataset of gears to apply to your network? The second question is, when I record a video of a gear moving, how do I transmit the video to the video segmentation network mentioned in your paper to achieve the output of spatio-temporal consistent posture? Looking forward to your reply, thank you again!