Related Concept Videos
Light Acquisition
Focusing of Light in the Eye
Flame Photometry: Overview
Group Polarization
Flame Photometry: Lab
Lampbrush Chromosomes
You might also read
Related Articles
Articles linked to this work by shared authors, journal, and citation graph.
Transformer-based arterial spin labeling perfusion MRI denoising.
3D objects reconstruction from frontal images: an example with guitars.
Local attention and long-distance interaction of rPPG for deepfake detection.
Related Experiment Video
Updated: Sep 22, 2025

Holistic Facial Composite Creation and Subsequent Video Line-up Eyewitness Identification Paradigm
Published on: December 24, 2015
Illumination-aware group portrait compositor.
Masaru Ohkawara1, Issei Fujishiro1
1Keio University, Yokohama, Japan.
This study introduces a new compositing framework for human figures, enhancing visual coherence with realistic lighting and shadows. The method effectively creates group portraits, fostering a sense of coexistence in digital environments.
Area of Science:
- Computer Graphics
- Computational Imaging
- Human-Computer Interaction
Background:
- Synthesizing realistic digital humans requires sophisticated compositing techniques.
- Existing methods often struggle to maintain surface details and integrate lighting realistically.
Purpose of the Study:
- To develop a novel compositing framework for full-length human figures.
- To achieve high visual coherence in composite images, including realistic light and shadow.
- To enable the creation of group portraits with a sense of coexistence.
Main Methods:
- A five-stage pipeline involving photometric information estimation, 3D reconstruction, global illumination simulation, and lighting transfer.
- Extending a previous compositing pipeline for enhanced functionality.
- Compositing separately rendered 3D human models.
Main Results:
- The framework successfully maintains surface details and integrates localized light and shadow.
- Composite results demonstrated high visual coherence, functioning effectively as a group portrait compositor.
- The proposed method outperformed rendering multiple avatars together in empirical evaluations.
Conclusions:
- The framework provides a new means for fostering a sense of coexistence in remote societies.
- It offers an efficient method for generating highly photorealistic cyberworlds.
- The approach advances the state-of-the-art in digital human rendering and compositing.

