acne-new Computer Vision Project

Jens

Updated 3 months ago

0

views

0

downloads
Classes (6)
acne
mole1
mole2
pimple
scar
spot
Description

Acne Dataset Fine-tuned:

  • This dataset includes 2436 images.
  • Annotations will be done by Amar and Vivek, each 1200 images.
  • Remaining 36 images, I will be labelling as examples.
  • Annotations are to be done for following mentioned classes ( 3 classes for first iteration).
  • Annotations are to be completed within 4 days of assignment, i.e. submission by July 11, 2023.
  • First batch uploaded on July 5, 2023.

For Iteration 1, annotate following classes:

0 - Acne : bumpy, small-big, white-brown cysts with no-redness, NOT RED.

1 - Pimple : bumpy, small-big, red-white cysts, any red big bump, MUST BE RED.

2 - Spots : flat, small, circular, dark brown.

For Iteration 2, annotate following classes:

3 - Mole1 : flat/black/small/concentrated (til type Moles)

4 - Mole2 : bumpy/black/dark-brown (Mussa type Moles)

5 - Scar : FLAT/ ANY SHAPE/ ANY SHADE

NOTE:

  1. Freckles: if too many freckles are present on the face, remove the image from the project, save it in a separate folder locally to be used later when need arise.
  2. Moles: if any moles are confusing and similar to spots definition, mark it as spot.
  3. Scars: for cuts/scars on face, if they are darker shade, circular, and small in size and look similar to spot, mark it as spot.

For now DO NOT annotate Moles, Bindi or Piercing, if we need, we add it later.

4 - Bindi : any small circular colored painting/sticker on the forehead center.

5 - Piercing : any peircing that is on the face, nose or eyebrows, or lips.

Supervision

Build Computer Vision Applications Faster with Supervision

Visualize and process your model results with our reusable computer vision tools.

Cite This Project

LICENSE
CC BY 4.0

If you use this dataset in a research paper, please cite it using the following BibTeX:

                        @misc{
                            acne-new-wzoih_dataset,
                            title = { acne-new Dataset },
                            type = { Open Source Dataset },
                            author = { Jens },
                            howpublished = { \url{ https://universe.roboflow.com/jens-r0xoc/acne-new-wzoih } },
                            url = { https://universe.roboflow.com/jens-r0xoc/acne-new-wzoih },
                            journal = { Roboflow Universe },
                            publisher = { Roboflow },
                            year = { 2024 },
                            month = { sep },
                            note = { visited on 2024-11-25 },
                            }
                        
                    

Similar Projects

See More