Mingxiao Tu

icon company University of Sydney icon location Sydney icon location PhD student at School of Computer Science, University of Sydney, Australia

Results 13 comments of Mingxiao Tu

Predictions contain a white rectangle in the middle while using my own data set

https://imgur.com/a/tAjhcvB here is the result of my own prediction mask

interesting video object segmentation (and tracking) result

Hi that is a really interesting result and thank you for sharing! May I ask how you get rid of the background (and all other occlusions) but only keep on...

Improve object tracking for objects that temporarily disappear and reappear in the video

thank you for the advice, I will give it a try!

export masks?

I am wondering will this code save a binary mask (with undifferentiable 0 and 1 values) to that png file? Shall I multiply with the original frame before saving?

Run time performance

Thank you for your reply! I have run it and the performance is really good, about 0.04s per frame in my case. Do you intend to release a demo code...

Saving SMPL Ground Truth and Point Cloud File

Hi Henry, I hope this message finds you well. Thank you very much for your prompt reply. Upon further reflection, I realize that I could have familiarized myself more thoroughly...

Does it support specific object segmentation ?

same question here, as I tried this on my video, but it seems that all objects in every frame will result in a different class, so I am wondering how...

Training problem

Hi Tiange, Thank you very much for your reply and the input. I changed the patch-size in the configs/occnerf/zju_mocap/387/occnerf.yaml from 32 to 16 first and will see how it goes....

Training problem

Hi Tiange, thank you very much for your input! I will close this issue then :)

Training problem

Sorry Tiange just one last question, if I'd like to further test the code on a customized data (for example, a 3D body scan with real occlusions, in a similar...