Activity
Nov
Dec
Jan
Feb
Mar
Apr
May
Jun
Jul
Aug
Sep
Oct
Nov
Mon
Wed
Fri
Ratings Progression
Loading...
Challenge Categories
Loading...
Challenges Entered
Understand semantic segmentation and monocular depth estimation from downward-facing drone images
Latest submissions
See Allgraded | 217632 | ||
graded | 217621 | ||
graded | 217190 |
5 Puzzles 21 Days. Can you solve it all?
Latest submissions
See Allgraded | 175319 | ||
graded | 175042 | ||
graded | 175037 |
Latest submissions
See Allgraded | 149268 | ||
graded | 149266 | ||
graded | 149264 |
Predict Mask & Bounding Box From Images
Latest submissions
See Allgraded | 175319 | ||
graded | 175042 | ||
graded | 175037 |
Perform semantic segmentation on aerial images from monocular downward-facing drone
Latest submissions
See Allgraded | 217632 | ||
graded | 217621 | ||
graded | 217190 |
Participant | Rating |
---|
Participant | Rating |
---|
Mask Prediction
anna_mrukwa has not provided any information yet.
Notebooks
-
U2Net Face deblurring using Nested U-Structure for Salient Object Detection; top1 - 0.798 SSIM scoreanna_mrukwaΒ· Over 2 years ago
The mistakes in the dataset
Over 2 years agoHello, I would like to ask about some of the issues with the dataset - after loading it again for further data inspection and noticing stark contrasts in the sizes of the masks in some of the photos, we noticed that some of them are possibly displaced - not covering the face or any part of it -
and really small, where the other ones are considerably distorted. The issue persists in all three parts of the dataset, but we may have not identified all of the problematic photos. In the test dataset, some of them are:
In the attachment to this topic I include a simple notebook presenting some of the images that are related to the issue and could be detected via simple analysis.
@Shubhamaicrowd