Aims and Scope: Analyzing the vehicle occupants’ faces and their dynamics in the complex driving environment is important in the design of driver assistance and active safety systems. The VIVA Face Analysis Challenge aims to engage the research community to participate and contribute towards development and evaluation of novel algorithms for face detection and head pose estimation. This challenge introduces a never before seen data set of looking at the occupants inside the vehicle and call for participants to take part in any or both parts of the challenge: face detection and head pose estimation.

The Challenge: The challenge is to robustly and accurately localize vehicle occupants’ faces and estimate their head pose under varying illumination (e.g. sunny and cloudy), in the presence of typical partially occluding objects (e.g. eyewear and hats) or actions (e.g. hand movements), under different camera configurations (e.g. camera perspective and resolution) and across different drivers. The dataset contains video frames from clips of naturalistic driving in highways and urban streets.

Submission: Participants can submit to any or both parts of the competition: face detection, head pose estimation. For each category, submissions will be ranked based on appropriate evaluations metrics.

Participation: If you are interested and planning to participate in the VIVA Face Challenge, please send the organizers an email at and visit respective challenge pages for more information: Face-Detection Challenge and Head-Pose Challenge

Important Dates:

Sujitha Martin, Kevan Yuen and Mohan Trivedi
University of California, San Diego, USA