Papers accepted!

We have four research papers accepted for publishing in major conferences and journals. (links coming soon!)

CZOVNY, R. ; BELLON, O. R. P. ; SILVA, L. ; COSTA, H.S.G. . Minutiae Matching using 3D Pore Clouds. In: IAPR International Conference on Pattern Recognition (ICPR), 2018, Pequim.

ALBIERO, V. ; BELLON, O. R. P. ; SILVA, L. . Multi-label action unit detection on multiple head poses with dynamic region learning. In: IEEE International Conference on Image Processing (ICIP), 2018, Atenas.

PARADA, D. M. P. ; ZAVAN, F. H. B. ; BELLON, O. R. P. ; SILVA, L. . Compound facial expressions in 3D faces: forced and spontaneous. In: 26th European Signal Processing Conference (EUSIPCO), 2018, Roma.
GOMES, L. ; BELLON, O. R. P. ; SILVA, L. . Exploring 3D reconstruction using RGB-D cameras for cultural heritage: A comparative approach and its application in Brazilian baroque sculptures. ACM Journal on Computing and Cultural Heritage, 2018.


Our students Flávio Zavan and Thiago Pradi will be presenting their master thesis at the Workshop of Theses and Dissertations (WTD). And Diandra Kubo her undergraduate work at the Workshop of Undergraduate Works (WUW).

We will also give a 6 hour advanced tutorial on face processing In-the-Wild. It will take place at SIBGRAPI  on Tuesday, October 17 starting at 9am.

Here is the abstract :

With the global demand for extra security systems, and the growing of human-machine
interaction, facial analysis in unconstrained environments (in the wild) became a hot-topic in
recent computer vision research. Unconstrained environments include surveillance footage,
social media photos and live broadcasts. This type of images and videos include no control
over illumination, position, size, occlusion, and facial expressions. Successful facial processing
methods for controlled scenarios are unable to pledge with challenging circumstances.
Consequently, methods tailored for handling those situations are indispensable for the face
analysis research progress. This work presents a comprehensive review of state-of-the-art
methods, drawing attention to the complications derived from in the wild scenarios and the
behavior differences when applied to the controlled images. The main topics to be covered
are: (1) face detection; (2) facial image quality; (3) head pose estimation; (4) face alignment;
(5) 3D face reconstruction; (6) gender and age estimation; (7) facial expressions and  emotions; and (8) face recognition. Finally, available code and applications for in the wild face analysis are presented, followed by a discussion on future directions.

Our paper ‘AUMPNet’ was accepted to FG FERA 2017

We had our paper ‘AUMPNet: simultaneous Action Units detection and intensity estimation on multipose facial images using a single convolutional neural network’ accepted for the FG FERA challenge 2017 edition, and our master students and authors Júlio César Batista and Vítor Albiero will be presenting the paper at Washington, USA between the days from may 30th to june 3rd.

Here is the challenge website :