Key Moments
TU Wien Rendering #21 - Tone Mapping Basics
Key Moments
Rendering challenge: Display large radiance ranges on limited screens. Tone mapping compresses & adapts.
Key Insights
Light simulations output radiance, but displays need RGB values.
Radiance values have a vast dynamic range (up to 80 million:1), far exceeding display capabilities (around 1000:1).
Human vision also has a large dynamic range (around 100 million:1), achieved through biochemical and neurological adaptation.
Tone mapping is essential to compress high dynamic range radiance to low dynamic range for display.
Tone mapping involves range compression and adherence to standardized color spaces.
Global tone mappers apply a uniform function, while local tone mappers consider pixel neighborhoods for adaptive contrast enhancement.
THE RADIANCE-TO-RGB CHALLENGE
Computer graphics simulations aim to calculate radiance, which quantifies light traveling through a scene. However, display devices like monitors cannot directly show this radiance. They expect RGB color values. While radiance can carry color information, converting the immense range of simulated radiance values into a format suitable for displays is a complex problem. This conversion is not straightforward due to the vast differences in light intensity found in the real world.
THE VAST DYNAMIC RANGE OF LIGHT
Real-world light intensities vary dramatically. For instance, the difference between sunlight and moonlight on Earth can be a factor of 800,000. Add to this the reflectivity of surfaces (white versus black), and the ratio between the darkest and brightest visible light in a single scene can reach an astonishing 80 million to one. This high dynamic range (HDR) is crucial for perceiving details in both bright and dark conditions, a capability evolved in human vision.
LIMITATIONS OF DISPLAY TECHNOLOGY
Human vision boasts a remarkable dynamic range, capable of adapting to light differences of up to 100 million to one through photochemical bleaching, pupil size, and neural adaptation. In contrast, standard displays have a significantly limited dynamic range, typically around 1,000 to one. Furthermore, 8-bit color encoding, common in image files, provides only 256 levels per color channel, further restricting the display of subtle gradations in brightness found in HDR scenes.
THE ROLE OF TONE MAPPING
Tone mapping, also known as tone reproduction, is the process designed to bridge the gap between the high dynamic range output of light simulations and the low dynamic range capabilities of display devices. It involves compressing the wide range of luminance values into a narrower range that can be effectively displayed without losing crucial visual information, ensuring that both dark and bright areas are perceivable.
APPROACHES TO TONE MAPPING
Tone mapping techniques can be broadly categorized into global and local methods. Global tone mappers apply a single, uniform mapping function across the entire image, making them computationally efficient and suitable for parallel processing on GPUs. Local tone mappers, however, consider the surrounding pixels of a given pixel, allowing for adaptive contrast enhancement. This local approach is perceptually motivated, mimicking how the human eye adapts its brightness perception based on local conditions.
GLOBAL VERSUS LOCAL MAPPING
Global tone mappers are fast and simpler to implement, processing each pixel independently with the same formula. This speed is advantageous for real-time rendering. However, they can suffer from a loss of detail because they don't account for local variations in scene luminance. Local tone mappers offer more sophisticated control by adjusting the mapping based on the pixel's neighborhood, enabling better detail preservation and contrast locally, but at the cost of increased computational complexity and slower performance.
DISTINGUISHING TRUE TONE MAPPING
It is crucial to differentiate technical tone mapping from artistic manipulation often seen in casual HDR photography. While tone mapping aims for a perceptually and physically validated representation of a scene's dynamic range, some HDR photography heavily overuses these capabilities, resulting in unnatural effects like excessive halos, oversaturated colors, and an overall artificial appearance. True tone mapping should produce realistic impressions, not aesthetic alterations for their own sake.
REINHARDT TONE MAPPER AND HDR PHOTOGRAPHY
The Reinhardt tone mapper is a commonly implemented algorithm in rendering software. It's robust enough to be applied in contexts like High Dynamic Range (HDR) photography, which involves combining multiple exposures of the same scene. When applied correctly, tone mapping in HDR photography should yield realistic results, subtly enhancing contrast and detail without the egregious visual artifacts associated with misused HDR techniques.
Mentioned in This Episode
●Software & Apps
●Concepts
Tone Mapping Dos and Don'ts
Practical takeaways from this episode
Do This
Avoid This
Common Questions
Tone mapping is a crucial technique in rendering and digital imaging used to compress the large dynamic range of luminance values from light simulations (high dynamic range) into a range that standard displays can reproduce (low dynamic range). This is necessary because displays have a much lower contrast ratio than the real world, and simply converting radiance to RGB would result in lost details (over/underexposure).
Topics
Mentioned in this video
A limitation in display technology where color channels are often encoded with 8 bits, resulting in a limited number of values (256 per channel) and an inadequate dark-to-bright ratio for realistic scenarios.
The standard color model expected by displays, which contrasts with the radiance values output by light simulations.
A digital photography technique that combines multiple exposure images of the same scene (taken with different exposure times) to capture a wider dynamic range. It often uses similar methodologies to tone mapping but can be misused artistically.
A method of ray tracing that traces rays carrying spectral radiance, not just radiance, allowing simulation of effects like light splitting into rainbow colors by prisms.
More from Two Minute Papers
View all 18 summaries
11 minThe Physics Bug That Stumped Everyone Is Finally Gone!
10 minAdobe & NVIDIA’s New Tech Shouldn’t Be Real Time. But It Is.
12 minThe Most Realistic Fire Simulation Ever
10 minNVIDIA’s Insane AI Found The Math Of Reality
Found this useful? Build your knowledge library
Get AI-powered summaries of any YouTube video, podcast, or article in seconds. Save them to your personal pods and access them anytime.
Try Summify free