[Ray Tracing] 1 An overview of ray tracing

Profile

I write this note to record general ideas and notable details to let me pick them up easily in the following procedure.

I decide to sort them after I find it is easy to forget the chapters you reads before.

1 Image synthesis

Introduce pinhole camera model and determination the average light (color) in one pixel.

2 Tracing rays

2.1 Forward rat tracing

follow photons from their origin at the light and into the scene, tracing their path in a forward direction, just as the photons themselves would have traveled it. unworkable: too much calculation, low computation efficiency

2.2 Backward ray tracing

Follow a ray backwards to find out where it may have begun.

note: ”first object hit by a ray” = ”the first object might have emitted that ray”

Term ”ray tracing” is often referred to backward ray tracing.

2.3 Ray combination

pixel ray/eye ray

illumination ray/shadow ray

reflection ray

transparent ray

2.4 ~ 2.7

These sections introduce the above concepts in detail.

Note:

1.Illumination ray or shadow ray are those coming from light source directly.

2. Don’t be confused of the literal meaning of incident ray. We are using backward ray tracing, so incident ray refers to rays into our eye or reflected from the surface.

3 Recursive visibility

Use recursive algorithm in ray tracing

3.1 Surface physics

Introduction of an eye ray propagated through a scene and a ray tree.

4 Aliasing common in signal processing

4.1 Spatial aliasing

cause jaggies with low sampling rate and resolution.

4.2 Temporal aliasing

1.Things are happening too fast to record accurately.

2. Small objects move across the screen and they will blink on or off, or pop.

3. Smoothly moving edges are another kind of pop.

4.3 Anti-aliasing

list some popular approaches to solve aliasing

4.4 ~ 4.7

Super sampling

adaptive super sampling

stochastic ray tracing

statistical ray tracing

The above approaches seem to relative with many algorithms in other fields. Maybe I can use some popular methods in signal processing, control theory and machine learning to get better performance in image.

Rendering equation

The rendering equation can express how light bounces around in a scene mathematically. It is useful and one powerful way to solve it is ray tracing.

Advertisements

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s