Christian Döring
Research Subjects
Christian Döring has a total of 14 co-authors. Here are the top co-authors by number of shared publications:
- Arno Coomans
Huawei Technologies Switzerland
3 shared publications - Edoardo Alberto Dominici
Huawei Technologies Switzerland
3 shared publications - Markus Steinberger
Graz University of Technology
3 shared publications - Floor Verhoeven
ETH Zurich, Switzerland
2 shared publications - Giacomo Nazzaro
Huawei Technologies
2 shared publications - Konstantinos Vardis
Department of Informatics, Athens University of Economics and Business, Athens, Greece
2 shared publications
Latest Publications
- Open Access
- 30 November 2026
- Published by Association for Computing Machinery (ACM)
- p. 1-12
Procedural Neural Shading of Mesoscale Detail
- 30 November 2026
- Published by Association for Computing Machinery (ACM)
- p. 1-12
Open AccessProcedural Neural Shading of Mesoscale Detail
- 15 April 2026
- Published by Wiley in Computer Graphics Forum
Real‐time Rendering with a Neural Irradiance Volume
Rendering diffuse global illumination in real‐time is often approximated by pre‐computing and storing irradiance in a 3D grid of probes. As long as most of the scene remains static, probes approximate irradiance for all surfaces immersed in the irradiance volume, including novel dynamic objects. This approach, however, suffers from aliasing artifacts and high memory consumption. We propose Neural Irradiance Volume (NIV), a neural‐based technique that allows accurate real‐time rendering of diffuse global illumination via a compact pre‐computed model, overcoming the limitations of traditional probe‐based methods, such as the expensive memory footprint, aliasing artifacts, and scene‐specific heuristics. The key insight is that neural compression creates an adaptive and amortized representation of irradiance, circumventing the cubic scaling of grid‐based methods. Our superior memory‐scaling improves quality by at least 10x at the same memory budget, and enables a straightforward representation of higher‐dimensional irradiance fields, allowing rendering of time‐varying or dynamic effects without requiring additional computation at runtime. Unlike other neural rendering techniques, our method works within strict realtime constraints, providing fast inference (around 1 ms per frame on consumer GPUs at full HD resolution), reduced memory usage (1–5 MB for medium‐sized scenes), and only requires a G‐buffer as input, without expensive ray tracing or denoising.
- 15 April 2026
- Published by Wiley in Computer Graphics Forum
Real‐time Rendering with a Neural Irradiance Volume
Rendering diffuse global illumination in real‐time is often approximated by pre‐computing and storing irradiance in a 3D grid of probes. As long as most of the scene remains static, probes approximate irradiance for all surfaces immersed in the irradiance volume, including novel dynamic objects. This approach, however, suffers from aliasing artifacts and high memory consumption. We propose Neural Irradiance Volume (NIV), a neural‐based technique that allows accurate real‐time rendering of diffuse global illumination via a compact pre‐computed model, overcoming the limitations of traditional probe‐based methods, such as the expensive memory footprint, aliasing artifacts, and scene‐specific heuristics. The key insight is that neural compression creates an adaptive and amortized representation of irradiance, circumventing the cubic scaling of grid‐based methods. Our superior memory‐scaling improves quality by at least 10x at the same memory budget, and enables a straightforward representation of higher‐dimensional irradiance fields, allowing rendering of time‐varying or dynamic effects without requiring additional computation at runtime. Unlike other neural rendering techniques, our method works within strict realtime constraints, providing fast inference (around 1 ms per frame on consumer GPUs at full HD resolution), reduced memory usage (1–5 MB for medium‐sized scenes), and only requires a G‐buffer as input, without expensive ray tracing or denoising.
- Open Access
Real-time Rendering with a Neural Irradiance Volume
Rendering diffuse global illumination in real-time is often approximated by pre-computing and storing irradiance in a 3D grid of probes. As long as most of the scene remains static, probes approximate irradiance for all surfaces immersed in the irradiance volume, including novel dynamic objects. This approach, however, suffers from aliasing artifacts and high memory consumption. We propose Neural Irradiance Volume (NIV), a neural-based technique that allows accurate real-time rendering of diffuse global illumination via a compact pre-computed model, overcoming the limitations of traditional probe-based methods, such as the expensive memory footprint, aliasing artifacts, and scene-specific heuristics. The key insight is that neural compression creates an adaptive and amortized representation of irradiance, circumventing the cubic scaling of grid-based methods. Our superior memory-scaling improves quality by at least 10x at the same memory budget, and enables a straightforward representation of higher-dimensional irradiance fields, allowing rendering of time-varying or dynamic effects without requiring additional computation at runtime. Unlike other neural rendering techniques, our method works within strict real-time constraints, providing fast inference (around 1 ms per frame on consumer GPUs at full HD resolution), reduced memory usage (1-5 MB for medium-sized scenes), and only requires a G-buffer as input, without expensive ray tracing or denoising.
Open AccessReal-time Rendering with a Neural Irradiance Volume
Rendering diffuse global illumination in real-time is often approximated by pre-computing and storing irradiance in a 3D grid of probes. As long as most of the scene remains static, probes approximate irradiance for all surfaces immersed in the irradiance volume, including novel dynamic objects. This approach, however, suffers from aliasing artifacts and high memory consumption. We propose Neural Irradiance Volume (NIV), a neural-based technique that allows accurate real-time rendering of diffuse global illumination via a compact pre-computed model, overcoming the limitations of traditional probe-based methods, such as the expensive memory footprint, aliasing artifacts, and scene-specific heuristics. The key insight is that neural compression creates an adaptive and amortized representation of irradiance, circumventing the cubic scaling of grid-based methods. Our superior memory-scaling improves quality by at least 10x at the same memory budget, and enables a straightforward representation of higher-dimensional irradiance fields, allowing rendering of time-varying or dynamic effects without requiring additional computation at runtime. Unlike other neural rendering techniques, our method works within strict real-time constraints, providing fast inference (around 1 ms per frame on consumer GPUs at full HD resolution), reduced memory usage (1-5 MB for medium-sized scenes), and only requires a G-buffer as input, without expensive ray tracing or denoising.
- 23 April 2024
- Published by Wiley in Computer Graphics Forum
- Vol. 43 (2)
Real‐time Neural Rendering of Dynamic Light Fields
Synthesising high‐quality views of dynamic scenes via path tracing is prohibitively expensive. Although caching offline‐quality global illumination in neural networks alleviates this issue, existing neural view synthesis methods are limited to mainly static scenes, have low inference performance or do not integrate well with existing rendering paradigms. We propose a novel neural method that is able to capture a dynamic light field, renders at real‐time frame rates at 1920×1080 resolution and integrates seamlessly with Monte Carlo ray tracing frameworks. We demonstrate how a combination of spatial, temporal and a novel surface‐space encoding are each effective at capturing different kinds of spatio‐temporal signals. Together with a compact fully‐fused neural network and architectural improvements, we achieve a twenty‐fold increase in network inference speed compared to related methods at equal or better quality. Our approach is suitable for providing offline‐quality real‐time rendering in a variety of scenarios, such as free‐viewpoint video, interactive multi‐view rendering, or streaming rendering. Finally, our work can be integrated into other rendering paradigms, e.g., providing a dynamic background for interactive scenarios where the foreground is rendered with traditional methods.
- 23 April 2024
- Published by Wiley in Computer Graphics Forum
- Vol. 43 (2)
Real‐time Neural Rendering of Dynamic Light Fields
Synthesising high‐quality views of dynamic scenes via path tracing is prohibitively expensive. Although caching offline‐quality global illumination in neural networks alleviates this issue, existing neural view synthesis methods are limited to mainly static scenes, have low inference performance or do not integrate well with existing rendering paradigms. We propose a novel neural method that is able to capture a dynamic light field, renders at real‐time frame rates at 1920×1080 resolution and integrates seamlessly with Monte Carlo ray tracing frameworks. We demonstrate how a combination of spatial, temporal and a novel surface‐space encoding are each effective at capturing different kinds of spatio‐temporal signals. Together with a compact fully‐fused neural network and architectural improvements, we achieve a twenty‐fold increase in network inference speed compared to related methods at equal or better quality. Our approach is suitable for providing offline‐quality real‐time rendering in a variety of scenarios, such as free‐viewpoint video, interactive multi‐view rendering, or streaming rendering. Finally, our work can be integrated into other rendering paradigms, e.g., providing a dynamic background for interactive scenarios where the foreground is rendered with traditional methods.
- 2 September 2010
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Computer Graphics and Applications
- Vol. 30 (6) , 29-41
Advanced Volume Illumination with Unconstrained Light Source Positioning
Advanced volume visualization techniques are important for both visual arts and visualization. An interactive volumetric lighting model simulates scattering and shadowing to generate high-quality images. Compared to gradient-based shading, this model achieves superior image quality and higher frame rates. User study results indicate that this model can also significantly improve depth perception.
- 2 September 2010
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Computer Graphics and Applications
- Vol. 30 (6) , 29-41
Advanced Volume Illumination with Unconstrained Light Source Positioning
Advanced volume visualization techniques are important for both visual arts and visualization. An interactive volumetric lighting model simulates scattering and shadowing to generate high-quality images. Compared to gradient-based shading, this model achieves superior image quality and higher frame rates. User study results indicate that this model can also significantly improve depth perception.
- Open Access
- 1 January 2004
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- Vol. 1, 84-89 Vol.1
Fuzzy clustering of quantitative and qualitative data
In many applications the objects to cluster are described by quantitative as well as qualitative features. A variety of algorithms has been proposed for unsupervised classification if fuzzy partitions and descriptive cluster prototypes are desired. However, most of these methods are designed for data sets with variables measured in the same scale type (only categorical, or only metric). We propose a new fuzzy clustering approach based on a probabilistic distance measure. Thus a major drawback of present methods can be avoided which ties in the vulnerability to favor one type of attributes.
- 1 January 2004
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- Vol. 1, 84-89 Vol.1
Open AccessFuzzy clustering of quantitative and qualitative data
In many applications the objects to cluster are described by quantitative as well as qualitative features. A variety of algorithms has been proposed for unsupervised classification if fuzzy partitions and descriptive cluster prototypes are desired. However, most of these methods are designed for data sets with variables measured in the same scale type (only categorical, or only metric). We propose a new fuzzy clustering approach based on a probabilistic distance measure. Thus a major drawback of present methods can be avoided which ties in the vulnerability to favor one type of attributes.