2016 Looking at People ICPR Challenge

Challenge description

---

Research advances in computer vision and pattern recognition have resulted in tremendous advances in different problems and applications. As a result, several problems on visual analysis can be considered as solved (e.g., face recognition), at least in certain scenarios and under specific circumstances.  Despite these important advances, there are still many open problems that are receiving much attention from the community because of the potential applications. We are organizing a contest around four of these problems that require, in addition to performing an effective visual analysis, to deal with multimodal information (e.g., audio,  RGB-D video, etc.) in order to be solved. In addition we focus on problems in which we aim to recognize non visually evident patterns (e.g., personality traits). The contest is supported by three organizations with vast experience and prestige in the organization of academic contests, namely: Chalearn, MediaEval and ImageCLEF. The contest is also supported by the IAPR TC 12 on visual and multimedia information systems. The ChaLearn Looking at People ICPR Contest 2016, to be held next December in conjunction with ICPR 2016 in Cancún, México.

There are four tracks associated to the contest:

1. First impressions track (Round 2). To recognize  apparent personality traits from 15-second videos, where big-five categories were considered.

2. Isolated and continuous gesture recognition. To recognize gestures in either segmented or continuous video, starting from RGB-D data and considering a large number of categories and domains.

3. Context of experience. To determine whether videos are suitable to be shown in a certain context.

 

Program

 

08:45h Opening: Presentation of the workshop, Hugo Jair Escalante (INAOE, Mexico) & Jun Wan (CAS, China)

09:00h Invited Speaker I: Towards machine interpretation of the real world. Alberto del Bimbo. (University of Florence, Italy)

Session chair: Hugo Jair Escalante

09:45h Session I: Challenge results presentation and award ceremony, Hugo Jair Escalante (INAOE, Mexico) & Jun Wan (CAS, China)

• ChaLearn Joint Contest on Multimedia Challenges Beyond Visual Analysis: An overview,  Hugo Jair Escalante, Víctor Ponce, Jun Wan., Michael Riegler, Albert Clapes, Sergio Escalera, Isabelle. Guyon, Xavier Baro, Paal Halvorsen, Henning Muller, and Martha Larson,

Session chair: Hugo Jair Escalante

10:00h Coffee Break

10:30h Session II: Winners First Impressions Challenge (Challenge Track 1)

• First impressions.

  • 1st: Multimodal Fusion of Audio, Scene, and Face Features for First Impression Estimation. Furkan Gürpınar, Heysem Kaya, Ali Salah

  • 4th Automatic Personality Prediction from Audiovisual Data using Random Forest Regression. Berkay Aydın, Ahmet Alp kindiroglu, Lale Akarun

Session chair: Hugo Jair Escalante

11:10h Session III: Winners Isolated Gesture Recognition Challenge (Challenge Track 2)

• 1st: Large-scale Gesture Recognition with a Fusion of RGB-D Data Based on C3D Model . Yunan Li, Qiguang Miao, Kuan Tian, Yingying Fan, Xin Xu, Rui Li, Jianfeng Song

• 2nd: Large-scale Isolated Gesture Recognition Using Convolutional Neural Networks. Pichao Wang, Wanqing Li, Song Liu, Zhimin Gao, Chang Tang, Philip Ogunbona,

• 3rd: Large-scale Isolated Gesture Recognition using Pyramidal 3D Convolutional Networks. Guangming Zhu, Liang Zhang, Lin Mei, Jie Shao, Juan Song, Peiyi Shen,

Session chair: Jun Wan

12:10h Lunch break (on your own)

13:30h Invited Speaker II: Multimodal Deep Learning. Fabio A. González O. (National University of Colombia, Colombia)

Session chair: Hugo Jair Escalante

14:15h Session IV: Affective computing session

• Predicting and visualizing psychological attributions with a deep neural network. Mariam Zabihi, Edward Grant, Stephan Sahm, Marcel Van Gerven

• Fusion of Classifier Predictions for Audio-Visual Emotion Recognition. Fatemeh Noroozi, Marina Marjanovic, Angelina Njegus, Sergio Escalera, Gholamreza Anbarjafari

Session chair: Jun Wan

15:00h Coffee Break

15:30h Invited Speaker III: Bayesian Classification: Applications in Computer Vision. L. Enrique Sucar (National Institute of Astrophysics, Optics and Electronics, Mexico)

Session chair: Hugo Jair Escalante

16:15h Session V: Winners Continuous Gesture Recognition Challenge (Challenge Track 3)

• • 1st: Two Streams Recurrent Neural Networks for Large-Scale Continuous Gesture Recognition.  Xiujuan Chai, Zhipeng Liu, Fang Yin, Zhuang Liu, Xilin Chen

  • 2nd: Using Convolutional 3D Neural Networks for User-Independent Continuous Gesture Recognition.Necati Camgoz, Simon Hadfield, Oscar Koller, Richard Bowden

  • 3rd: Large-scale Continuous Gesture Recognition Using Convolutional Neural Networks. Pichao Wang, Wanqing Li, Song Liu, Yuyao Zhang, Zhimin Gao, Philip Ogunbona

Session chair: Jun Wan

17:15h Closing: Closing ceremony & announcements,Hugo Jair Escalante (INAOE, Mexico) & Jun Wan (CAS, China)