Topics

This year's HBU workshop, in addition to covering the main themes of human behavior understanding, specifically deals with generating visual data of human behavior.

This workshop will solicit human behavior analysis solutions that clearly advance the field, and also to propose novel application scenarios. The covered topics may span items from each of the following topic dimensions, as well as target a focus theme challenge.

More precisely we seek for contributions describing new methodologies allowing to generate either still images or videos containing face(s) and/or full body(ies) or modifying attributes of faces (e.g. expression) and/or bodies (e.g. human pose). Contributions based on deep neural architectures are welcome, as well as methods based on parametric models.

Human Behavior Analysis Systems

• Action and activity recognition
• Affect analysis
• Face analysis
• Gaze, attention and saliency
• Gestures and haptic interaction
• Learning and adaptation
• Social signal processing
• Voice and speech analysis

Theory and Methodology of Human Behavior

• Theoretical frameworks of behavior analysis
• Data collection, annotation, and benchmarking
• Interaction design
• Theoretical frameworks of behavior analysis
• User studies and human factors

Generating Visual data of Human Behavior

• Generation of faces and modification of their attributes (e.g. age, expression).
• Generation of human bodies for behavior analysis.
• Synthesis of visual data representing individual and collective human activities.
• Deep generative models for image/video generation of human behavior.
• Parametric models for the synthesis of faces/bodies.
• Generation of multi-modal (vision and beyond) human behavioral data.
• Exploration of text-based descriptions linked with visual representation generation.
• Applications (e.g. surveillance, autonomous driving, fashion, robotics).