Pete Skomoroch of Datawrangling admits that work and other commitments have made him cut down on blogging significantly. However every now and then, he comes up with posts that literally show the astonishing amounts of sustained efforts put in by him.
Though the posting frequency is very low, It still qualifies as one of my favorite blogs simply because it is of great help to me.
Last week, he posted a notification that he had updated his old post which indexed a very large number of datasets on various topics/fields/projects. This post is an absolute life saver.
Please follow this post for an incredible list of datasets:
Click above to follow
Whenever I have searched for datasets for my problem/instructional/experimental requirements, I have almost always landed on Pete’s page.
Please check that link out for sure! I maintain a list largely constructed using the Data-Wrangling blog for face recognition. And since I have written two posts on face recognition before this one, it makes perfect sense to post that list.
Some Datasets for Face Recognition/Authentication/Detection purposes:
6. The AR Face Database (126 People, over 4000 color images, different illumination conditions, facial expressions and occlusions, two sessions per person)
7. Olivetti Research Limited (a database of about 400 images).
8. University of Berne, Switzerland Face Database (Frontal images of 30 people, 10 images for each with different orientation. Profile images of the same 30 people, 5 images for each).
9. University of Oulu – Physics Face Database (125 faces in 16 different illumination and camera calibration conditions, additional 16 if the person wears glasses).
10. The Georgia Tech Face Database (Images of 50 different subjects taken over two three sessions)
11. The Yale Database (One of the most widely used! 165 grayscale images of 15 indivduals)
12. The Yale Database B (5760 single light source images of 10 subjects in 576 viewing conditions )
13. Labeled Faces in the Wild (U-Mass, For the problem of unconstrained facial recognition. With 13000 images collected from the web, two different images for some individuals)
15. University of Sheffield Face Database (564 images of 20 individuals, mixed race, gender and appearance)
16. University of Essex Face Database (total 7900 images of 395 individuals with 20 images each )
17. Indian Face Database (Collected at IIT-K, 40 subejcts with 11 different poses)
19. NIST Mugshots
20. Face in Action (FIA) video Database (CMU)
21. The MIT-CBCL Database (Used in a previous post by me. Face Images of 10 subject. Huge and rather simplistic database )
23. Facial Expression Database (Cohn-Kanade-AU-Coded Facial Expression Database. Image data consists of 500 image sequences from 100 subjects)
24. AT&T Database (10 different images each of 40 different subjects)
26. BJUT 3D Face Database (500 individuals – 250 males and 250 females, 3-D Face Data)
29. CAS-PEAL Face Database (99,594 images if 1040 individuals, 595 males and 445 females with varying pose, expression, accessory and lighting)
30. VALID Database
33. 3D_RMA (3-D Face Image Database)
34. Equinox (Human Identification at a Distance)
I would try to edit this post when I get the time and insert details for each database to facilitate ease in navigation. I apologize for not doing so right away, I did not do so as it is a very time consuming process.
PS: I was tempted to re-blog the list by Pete, but I decided against it. It is his work, he deserves all the kudos!