-
I've reset the pipeline and run it again with all traceur fabmob data, which is not so much,
my user for example has 30109 background locations, which is among the highest number.
The computation t…
-
Hello,
After going through your data, you just labeled objects which has boxes. The background like sky or water are not labeled.
Therefore, I am curious how your data can be used for semantic s…
-
Hi!
I have a question about background in the calculation of the mIoU.
When we calculate the mIoU of the segmentaion, usually IoU of the background is included.
In that case, mIoU is at leas…
-
The HDRs on the NAS should be gone through-
- [x] Add some naming so we know what the HDR is of (like is done for the Montreal HDRs).
- [ ] The segmentation masks should be fixed up - there are …
-
Once we have the "slide preview" activated, the bar should be 50-60% transparent. Unless it already is.
-
Hi guys, thank for the great work! I have integrated my own segmentation and landmark detection model in phase 1, everything working really smooth. But now i have faced challenge when trying to integr…
-
Can I finetune the model using multiple bounding boxes for the same Image
(eg. in case of KITTI Dataset if I want to finetune the model to segment cars from background and there are multiple instanc…
-
If you're interested in doing further training, deeppomf has an image dataset scraped from yiff.party (includes NSFW images) where PSDs are used to generate pixel-perfect segmentations of figures: htt…
gwern updated
4 years ago
-
For semantic segmentation work, it would be nice to be able to specify image backgrounds and have the serialized data overlay those images as colored masks, maybe with variable opacity.
-
Thanks for the wonderful work. I have a question, apologies if I have missed the information.
As I understand, SAM is class-agnostic. Is it correct to assume that SAM does not support multi-objec…
MLDeS updated
11 months ago