Lecture 9: Intro to Ray-Tracing & Accelerating Ray-Scene Intersection (6)
stexus
What does it mean to cast a ray towards the light here? Is it the eye ray that was mentioned earlier?
lwg0320
@stexus In general, it means draw a line from the point to the light source to test if the light will shine on that point.
adityasingh7311
I don't fully understand what checking for shadows means computationally. How do we check for "collisions" of the light ray to detect a shadow? Latter slides describe how the find the ray-surface intersections but how do we know which objects/surfaces to check for?
edit: ok so slide 26 does answer my question I think. Exhaustive search with every object does seem to be the most basic approach but there are acceleration techniques.
Staffjamesfong1
@adityasingh7311 To answer the first question, checking for shadows just means we want to know whether or not a photon can go directly from the light source onto the surface in question. In this simple ray-casting scenario, the only way for a surface to be illuminated by a light is for it to pass the shadow check.
LucasArmand
It's interesting how the time-reversibility of physics allows us to cast light backwards and still generate physically accurate results. Is it even possible for this implementation to fail at simulating some feature due to it's reversed nature?
What does it mean to cast a ray towards the light here? Is it the eye ray that was mentioned earlier?
@stexus In general, it means draw a line from the point to the light source to test if the light will shine on that point.
I don't fully understand what checking for shadows means computationally. How do we check for "collisions" of the light ray to detect a shadow? Latter slides describe how the find the ray-surface intersections but how do we know which objects/surfaces to check for?
edit: ok so slide 26 does answer my question I think. Exhaustive search with every object does seem to be the most basic approach but there are acceleration techniques.
@adityasingh7311 To answer the first question, checking for shadows just means we want to know whether or not a photon can go directly from the light source onto the surface in question. In this simple ray-casting scenario, the only way for a surface to be illuminated by a light is for it to pass the shadow check.
It's interesting how the time-reversibility of physics allows us to cast light backwards and still generate physically accurate results. Is it even possible for this implementation to fail at simulating some feature due to it's reversed nature?