How can I make an animation?

[Nanran]

Hello！I am interested in your project name raytracer on github. I wonder how to generate "animation.gif", I can not find the source code. Could you please help me out?Thank you! I would be appreciate if you could send your all source code to me.

[rafael-fuente]

Hi! The method that can be used to make animations is create_animation. You can found it in animation.py

This method takes as an argument the scene you want to animate, and a function named update_scene that tells how the scene is changed every frame. This method will create an image for each frame that later you can convert into a video using FFmpeg

Here is a script that animates example1.py

from sightpy import *


# First, we set up the scene to animate like in example1.py:

# define materials to use
gold_metal = Glossy(diff_color = rgb(1., .572, .184), n = vec3(0.15+3.58j, 0.4+2.37j, 1.54+1.91j), roughness = 0.0, spec_coeff = 0.2, diff_coeff= 0.8) # n = index of refraction
bluish_metal = Glossy(diff_color = rgb(0.0, 0, 0.1), n = vec3(1.3+1.91j, 1.3+1.91j, 1.4+2.91j), roughness = 0.2,spec_coeff = 0.5, diff_coeff= 0.3)

floor =  Glossy(diff_color = image("checkered_floor.png", repeat = 80.),   
	            n = vec3(1.2+ 0.3j, 1.2+ 0.3j, 1.1+ 0.3j), roughness = 0.2, spec_coeff = 0.3, diff_coeff= 0.9 )


# Set Scene 
Sc = Scene(ambient_color = rgb(0.05, 0.05, 0.05))


angle = -np.pi/2 * 0.3
Sc.add_Camera(look_from = vec3(2.5*np.sin(angle), 0.25, 2.5*np.cos(angle)  -1.5 ), 
			  look_at = vec3(0., 0.25, -3.), 
	          screen_width = 400 ,
	          screen_height = 300)



Sc.add_DirectionalLight(Ldir = vec3(0.52,0.45, -0.5),  color = rgb(0.15, 0.15, 0.15))


Sc.add(Sphere(material = gold_metal, center = vec3(-.75, .1, -3.),radius =  .6, max_ray_depth = 3))
Sc.add(Sphere(material = bluish_metal, center = vec3(1.25, .1, -3.), radius = .6, max_ray_depth = 3))

Sc.add(Plane(material = floor,  center = vec3(0, -0.5, -3.0), width = 120.0,height = 120.0, u_axis = vec3(1.0, 0, 0), v_axis = vec3(0, 0, -1.0),  max_ray_depth = 3))

#see sightpy/backgrounds
Sc.add_Background("stormydays.png")


# Now we define the method to make the animation:

def update_scene(scene , time):
   #this function updates the scene every frame

   # update camera position
    angle = -np.pi/2 * 0.3 + 0.1*time
    scene.camera.look_from = vec3(4.0*np.sin(angle), 0.25, 4.0*np.cos(angle)  - 3. )

   # update camera orientation
    scene.camera.cameraFwd = (scene.camera.look_at - scene.camera.look_from).normalize()
    scene.camera.cameraRight = (scene.camera.cameraFwd.cross(vec3(0.,1.,0.))).normalize()
    scene.camera.cameraUp = scene.camera.cameraRight.cross(scene.camera.cameraFwd)


#Finally we render the animation:

create_animation(scene = Sc, 
	         samples_per_pixel = 5, 
		 fps = 5,        #frames per second
	         start_time = 0, #starting second
                 final_time = 4, #total seconds
	         update_scene = update_scene, #function that update the scene every frame 
	         name = "example1_frame") #name of frames

This script creates a folder named "frames" that contains all frames of the animation.
To convert them into a video run from the "frames" folder with the command prompt:
ffmpeg -r 5 -f image2 -s 854x480 -i example1_frame_%d.png -vcodec libx264 -crf 1 -pix_fmt yuv420p example1_animated.mp4

The video generated with this code is 5fps. For better quality use at least 30fps, although the rendering will take longer.

[AlkisPis]

Good, but a duration of only 4 secs --even 10 secs-- creates an annoying GIF-type video. I have created a video of one min in which the annoying GIF-type looping is hardly felt. But then this requires really a lot of frames which are useless after creating an MP4 from them.
The second animation method --'create_animation_using_opencv()'-- which creates the MP4 file directly is much more effective.

[AlkisPis]

I wonder if the 'update_scene()' method can be used with OpenCV or pyglet to create a live animation, w/o the need to create a video ...

[Nanran]

Thank you so much for your help! I have successfully gained the expected results.