Back to search

Athens University Economics and Business (AUEB)

Efficient algorithms for inverse lighting design

Abstract

dc:description

Lighting design is a demanding but essential process in computer cinematography, games, architectural design and various other applications for correctly illuminating or highlighting parts of a scene and enhancing storytelling. When targeting specific illumination goals and constraints, this process can be tedious, time-consuming and counter-intuitive even for experienced users and thus automatic, goal-driven methods are necessary for the estimation of a lighting configuration that generates the desired result. The goal of this thesis is to improve the pipeline of the inverse lighting design process. As a first step, we focus on making the illumination evaluation - a central part of our cost function - more efficient and robust. Aside from the inverse lighting design problem, the pursue of a highly-efficient global illumination technique is important for other major computer graphics applications, such as photorealistic real-time rendering. To this end, we present a generic approach that improves the efficiency of volume-based global illumination methods, using radiance field compression, cache point population optimization and a view-independent approximate indirect occlusion estimation. Next, we investigate and propose two methods that solve the generic inverse lighting design problem of light placement, focusing on scenes with high geometrical complexity, given a set of user-specified lighting goals. Our approach is based on a layered linear / non-linear optimization strategy and the introduction of a special light source indexing according to the compatibility of each individual luminary position with the desired illumination. To additionally minimize the number of used light sources, we offer a method that employs a special hierarchical light clustering that operates in the lighting goal coverage domain and overcomes limitations of previous approaches in environments with high occlusion or structural complexity. Both approaches are independent of the underlying light transport model and can quickly converge to usable configuration recommendations. In the context of perceptually-driven illumination, we propose methods that accentuate the prominent geometric features of objects by enhancing the local contrast near the latter. We present a method that computes a configuration of directional light sources that is based on a voting mechanism and a clustering that produces consistent and view-independent results. Expanding this idea and aiming at more practical constraints for physical environments, such as exhibition and museum lighting solutions, we present a method for the automatic arrangement of spot lights and other luminaires on or near user-provided arbitrary mounting surfaces. Our technique generates the most appropriate position and orientation for light sources based on a local contrast maximization near salient geometric feature groups and a clustering mechanism for determining the number of luminaires.

Degree

thesis:*
Grantor dc:publisher
Athens University Economics and Business (AUEB)
Year dc:date
2019

Author and committee

dc:creator, dc:contributor.*
Authors dc:creator
  • Gkaravelis, Anastasios
  • Γκαραβέλης, Αναστάσιος

Subjects

dc:subject × 12

Rights

dc:rights
Statement dc:rights
  • BY
Language dc:language
eng

Identifiers

dc:identifier.*
Identifier
10.12681/eadd/46846

Chain of custody

source
Harvested from
Greek National Archive of PhD Theses
Base URL
phdtheses.ekt.gr/eadd_oai/request
Last updated
2026-11-18
Source record
OAI-PMH GetRecord
citation

Gkaravelis, Anastasios; Γκαραβέλης, Αναστάσιος. Efficient algorithms for inverse lighting design. Athens University Economics and Business (AUEB), 2019. http://hdl.handle.net/10442/hedi/46846