search

pushkalkatara / DeepMPCVS

Deep Model Predictive Control For Visual Servoing, CoRL-2020.
https://corlconf.github.io/corl2020/paper_448/
21 stars 2 forks source link

How is the habitat env loaded? #2

Open jayaramreddy10 opened 2 years ago

jayaramreddy10 commented 2 years ago

Hi Pushkal,

1.) I want to know how are you loading habitat env say quantico in below code, is there a glb file specific to that, if yes can you point to that .glb file?

class HabitatEnv(): def init(self, folder, init_state, depth_type): scene_glb = folder + "/" + os.path.basename(folder).capitalize() + ".glb"

2.) Also, you are giving initial, final pose images as below. So, if I want to test on quantico env, above scene_glb should also be quantico? Do the below and above snippets need to have same scene?

f = open(folder + "/log.txt","w+")
f_pe = open(folder + "/photo_error.txt", "w+")
f_pose = open(folder + "/pose.txt", "w+")
img_source_path = folder + "/results/" + "test.rgba.00000.00000.png"
img_goal_path = folder + "/des.png"
img_src = read_gen(img_source_path)
img_goal = read_gen(img_goal_path)
d1 = plt.imread(folder + "/results/" + "test.depth.00000.00000.png")

3.) Now, if we want to test on random scene (by giving our own init.png, des.png. What should be scene_glb in def init of class HabitatEnv?