We present a novel precomputed radiance transfer method for efficient relighting under all-frequency environment illumination. Environment illumination is represented as a set of environment lights. Each environment light comprises a direction and an intensity. In a preprocessing step, the environment lights are clustered into several clusters, taking into account only the light directions. By experiment, we confirmed that the environment lights can be clustered into a much smaller number of clusters than their original number. Given any environment illumination, sampled as an environment map, an efficient relighting is then achieved by computing the radiance using the precomputed clusters. The proposed method enables relighting under very high-resolution environment illumination. In addition, unlike previous approaches, the proposed method can efficiently perform relighting when some regions of the given environment illumination change.
The copyright of the original papers published on this site belongs to IEICE. Unauthorized use of the original or translated papers is prohibited. See IEICE Provisions on Copyright for details.
Copy
Henry JOHAN, Tomoyuki NISHITA, "Clustering Environment Lights for an Efficient All-Frequency Relighting" in IEICE TRANSACTIONS on Information,
vol. E89-D, no. 9, pp. 2562-2571, September 2006, doi: 10.1093/ietisy/e89-d.9.2562.
Abstract: We present a novel precomputed radiance transfer method for efficient relighting under all-frequency environment illumination. Environment illumination is represented as a set of environment lights. Each environment light comprises a direction and an intensity. In a preprocessing step, the environment lights are clustered into several clusters, taking into account only the light directions. By experiment, we confirmed that the environment lights can be clustered into a much smaller number of clusters than their original number. Given any environment illumination, sampled as an environment map, an efficient relighting is then achieved by computing the radiance using the precomputed clusters. The proposed method enables relighting under very high-resolution environment illumination. In addition, unlike previous approaches, the proposed method can efficiently perform relighting when some regions of the given environment illumination change.
URL: https://global.ieice.org/en_transactions/information/10.1093/ietisy/e89-d.9.2562/_p
Copy
@ARTICLE{e89-d_9_2562,
author={Henry JOHAN, Tomoyuki NISHITA, },
journal={IEICE TRANSACTIONS on Information},
title={Clustering Environment Lights for an Efficient All-Frequency Relighting},
year={2006},
volume={E89-D},
number={9},
pages={2562-2571},
abstract={We present a novel precomputed radiance transfer method for efficient relighting under all-frequency environment illumination. Environment illumination is represented as a set of environment lights. Each environment light comprises a direction and an intensity. In a preprocessing step, the environment lights are clustered into several clusters, taking into account only the light directions. By experiment, we confirmed that the environment lights can be clustered into a much smaller number of clusters than their original number. Given any environment illumination, sampled as an environment map, an efficient relighting is then achieved by computing the radiance using the precomputed clusters. The proposed method enables relighting under very high-resolution environment illumination. In addition, unlike previous approaches, the proposed method can efficiently perform relighting when some regions of the given environment illumination change.},
keywords={},
doi={10.1093/ietisy/e89-d.9.2562},
ISSN={1745-1361},
month={September},}
Copy
TY - JOUR
TI - Clustering Environment Lights for an Efficient All-Frequency Relighting
T2 - IEICE TRANSACTIONS on Information
SP - 2562
EP - 2571
AU - Henry JOHAN
AU - Tomoyuki NISHITA
PY - 2006
DO - 10.1093/ietisy/e89-d.9.2562
JO - IEICE TRANSACTIONS on Information
SN - 1745-1361
VL - E89-D
IS - 9
JA - IEICE TRANSACTIONS on Information
Y1 - September 2006
AB - We present a novel precomputed radiance transfer method for efficient relighting under all-frequency environment illumination. Environment illumination is represented as a set of environment lights. Each environment light comprises a direction and an intensity. In a preprocessing step, the environment lights are clustered into several clusters, taking into account only the light directions. By experiment, we confirmed that the environment lights can be clustered into a much smaller number of clusters than their original number. Given any environment illumination, sampled as an environment map, an efficient relighting is then achieved by computing the radiance using the precomputed clusters. The proposed method enables relighting under very high-resolution environment illumination. In addition, unlike previous approaches, the proposed method can efficiently perform relighting when some regions of the given environment illumination change.
ER -