Spaces:
Runtime error
Runtime error
Update visualization/logger.py
Browse files- visualization/logger.py +82 -86
visualization/logger.py
CHANGED
|
@@ -4,20 +4,28 @@ from typing import Dict, Any, List
|
|
| 4 |
from utils.geometry import vector3_to_numpy, euler_to_quaternion
|
| 5 |
|
| 6 |
|
| 7 |
-
def create_subject_box(frame: Dict, dimensions: Dict) -> Dict[str, np.ndarray]:
|
|
|
|
|
|
|
|
|
|
|
|
|
| 8 |
position = vector3_to_numpy(frame['position'])
|
| 9 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
| 10 |
|
| 11 |
return {
|
| 12 |
'center': position,
|
| 13 |
-
'half_size':
|
| 14 |
}
|
| 15 |
|
| 16 |
|
| 17 |
class SimulationLogger:
|
| 18 |
def __init__(self):
|
| 19 |
rr.init("camera_simulation")
|
| 20 |
-
rr.log("world", rr.ViewCoordinates.RIGHT_HAND_Y_UP)
|
| 21 |
|
| 22 |
self.K = np.array([
|
| 23 |
[500, 0, 960],
|
|
@@ -29,52 +37,72 @@ class SimulationLogger:
|
|
| 29 |
if not instructions:
|
| 30 |
return
|
| 31 |
|
| 32 |
-
rr.log(
|
| 33 |
-
"
|
| 34 |
-
|
| 35 |
-
|
| 36 |
-
|
| 37 |
-
|
| 38 |
-
|
| 39 |
-
|
| 40 |
-
|
| 41 |
-
|
| 42 |
-
|
| 43 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
| 44 |
|
| 45 |
-
def log_subjects(self, subjects: List[Dict[str, Any]]
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 46 |
if not subjects:
|
| 47 |
return
|
| 48 |
|
| 49 |
-
|
| 50 |
-
|
| 51 |
-
|
| 52 |
-
|
|
|
|
| 53 |
|
| 54 |
-
|
| 55 |
-
|
| 56 |
-
# Use the frame at frame_idx, or the first frame if frame_idx is out of range
|
| 57 |
-
frame = subject['frames'][min(frame_idx, len(subject['frames']) - 1)]
|
| 58 |
-
box_params = create_subject_box(frame, subject['dimensions'])
|
| 59 |
-
centers.append(box_params['center'])
|
| 60 |
-
half_sizes.append(box_params['half_size'])
|
| 61 |
-
colors.append([0.8, 0.2, 0.2, 1.0])
|
| 62 |
-
labels.append(subject.get('objectClass', 'Unknown'))
|
| 63 |
-
except Exception as e:
|
| 64 |
-
print(f"Error creating box parameters: {str(e)}")
|
| 65 |
-
continue
|
| 66 |
|
| 67 |
-
|
| 68 |
-
|
| 69 |
-
|
| 70 |
-
|
| 71 |
-
|
| 72 |
-
|
| 73 |
-
|
| 74 |
-
|
| 75 |
-
|
| 76 |
-
|
| 77 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 78 |
|
| 79 |
def log_camera_trajectory(self, camera_frames: List[Dict[str, Any]]) -> None:
|
| 80 |
if not camera_frames:
|
|
@@ -88,36 +116,29 @@ class SimulationLogger:
|
|
| 88 |
"world/camera_trajectory",
|
| 89 |
rr.Points3D(
|
| 90 |
camera_positions,
|
| 91 |
-
colors=np.full((len(camera_positions), 4),
|
| 92 |
-
|
| 93 |
-
|
| 94 |
)
|
| 95 |
|
| 96 |
if len(camera_positions) > 1:
|
| 97 |
-
lines = np.stack(
|
| 98 |
-
[camera_positions[:-1], camera_positions[1:]], axis=1)
|
| 99 |
rr.log(
|
| 100 |
"world/camera_trajectory/line",
|
| 101 |
rr.LineStrips3D(
|
| 102 |
lines,
|
| 103 |
colors=[(0.0, 0.8, 0.8, 1.0)]
|
| 104 |
-
)
|
|
|
|
| 105 |
)
|
| 106 |
|
| 107 |
except Exception as e:
|
| 108 |
print(f"Error logging camera trajectory: {str(e)}")
|
| 109 |
|
| 110 |
-
def log_camera_frames(self, camera_frames: List[Dict[str, Any]]
|
| 111 |
-
"""Log camera frames and optionally their corresponding subjects"""
|
| 112 |
-
if subjects is None:
|
| 113 |
-
subjects = []
|
| 114 |
if not camera_frames:
|
| 115 |
return
|
| 116 |
|
| 117 |
-
# First log the camera trajectory
|
| 118 |
-
self.log_camera_trajectory(camera_frames)
|
| 119 |
-
|
| 120 |
-
# Then log each frame with its corresponding subject positions
|
| 121 |
for frame_idx, camera_frame in enumerate(camera_frames):
|
| 122 |
try:
|
| 123 |
rr.set_time_sequence("frame_idx", frame_idx)
|
|
@@ -142,9 +163,6 @@ class SimulationLogger:
|
|
| 142 |
)
|
| 143 |
)
|
| 144 |
|
| 145 |
-
# Log subjects for this frame
|
| 146 |
-
self.log_subjects(subjects, frame_idx)
|
| 147 |
-
|
| 148 |
except Exception as e:
|
| 149 |
print(f"Error logging camera frame {frame_idx}: {str(e)}")
|
| 150 |
|
|
@@ -161,32 +179,10 @@ class SimulationLogger:
|
|
| 161 |
helper_positions,
|
| 162 |
radii=np.full(len(helper_positions), 0.03),
|
| 163 |
colors=np.full((len(helper_positions), 4), [1.0, 1.0, 0.0, 1.0]),
|
| 164 |
-
)
|
|
|
|
| 165 |
)
|
| 166 |
|
| 167 |
for keyframe_idx, helper_keyframe in enumerate(helper_keyframes):
|
| 168 |
try:
|
| 169 |
-
|
| 170 |
-
rotation_q = euler_to_quaternion(helper_keyframe['angle'])
|
| 171 |
-
|
| 172 |
-
rr.log(
|
| 173 |
-
f"world/helper_camera_{keyframe_idx}",
|
| 174 |
-
rr.Transform3D(
|
| 175 |
-
translation=position,
|
| 176 |
-
rotation=rr.Quaternion(xyzw=rotation_q),
|
| 177 |
-
scale=(.5, .5, .5)
|
| 178 |
-
)
|
| 179 |
-
)
|
| 180 |
-
|
| 181 |
-
rr.log(
|
| 182 |
-
f"world/helper_camera_{keyframe_idx}/image",
|
| 183 |
-
rr.Pinhole(
|
| 184 |
-
image_from_camera=self.K,
|
| 185 |
-
width=1920,
|
| 186 |
-
height=1080,
|
| 187 |
-
)
|
| 188 |
-
)
|
| 189 |
-
|
| 190 |
-
except Exception as e:
|
| 191 |
-
print(
|
| 192 |
-
f"Error logging helper keyframe {keyframe_idx}: {str(e)}")
|
|
|
|
| 4 |
from utils.geometry import vector3_to_numpy, euler_to_quaternion
|
| 5 |
|
| 6 |
|
| 7 |
+
def create_subject_box(frame: Dict[str, Any], dimensions: Dict[str, float]) -> Dict[str, np.ndarray]:
|
| 8 |
+
"""
|
| 9 |
+
Given a single frame (position, rotation) and subject's dimensions,
|
| 10 |
+
return the center and half_size for a box representation.
|
| 11 |
+
"""
|
| 12 |
position = vector3_to_numpy(frame['position'])
|
| 13 |
+
half_size = np.array([
|
| 14 |
+
dimensions['width'] / 2.0,
|
| 15 |
+
dimensions['height'] / 2.0,
|
| 16 |
+
dimensions['depth'] / 2.0
|
| 17 |
+
], dtype=np.float32)
|
| 18 |
|
| 19 |
return {
|
| 20 |
'center': position,
|
| 21 |
+
'half_size': half_size
|
| 22 |
}
|
| 23 |
|
| 24 |
|
| 25 |
class SimulationLogger:
|
| 26 |
def __init__(self):
|
| 27 |
rr.init("camera_simulation")
|
| 28 |
+
rr.log("world", rr.ViewCoordinates.RIGHT_HAND_Y_UP, timeless=True)
|
| 29 |
|
| 30 |
self.K = np.array([
|
| 31 |
[500, 0, 960],
|
|
|
|
| 37 |
if not instructions:
|
| 38 |
return
|
| 39 |
|
| 40 |
+
rr.log(
|
| 41 |
+
"metadata/instructions",
|
| 42 |
+
rr.TextDocument(
|
| 43 |
+
"\n".join([
|
| 44 |
+
f"Instruction {i+1}:\n"
|
| 45 |
+
f" Movement: {inst.get('cameraMovement', 'N/A')}\n"
|
| 46 |
+
f" Easing: {inst.get('movementEasing', 'N/A')}\n"
|
| 47 |
+
f" Frames: {inst.get('frameCount', 'N/A')}\n"
|
| 48 |
+
f" Camera Angle: {inst.get('initialCameraAngle', 'N/A')}\n"
|
| 49 |
+
f" Shot Type: {inst.get('initialShotType', 'N/A')}\n"
|
| 50 |
+
f" Subject Index: {inst.get('subjectIndex', 'N/A')}"
|
| 51 |
+
for i, inst in enumerate(instructions)
|
| 52 |
+
])
|
| 53 |
+
),
|
| 54 |
+
timeless=True
|
| 55 |
+
)
|
| 56 |
|
| 57 |
+
def log_subjects(self, subjects: List[Dict[str, Any]]) -> None:
|
| 58 |
+
"""
|
| 59 |
+
Updated to handle dynamic subjects:
|
| 60 |
+
- Each subject has multiple frames (position, rotation).
|
| 61 |
+
- Subject dimensions (width, height, depth) remain constant across frames.
|
| 62 |
+
"""
|
| 63 |
if not subjects:
|
| 64 |
return
|
| 65 |
|
| 66 |
+
# For each subject, log each frame at a different time index.
|
| 67 |
+
for subject_idx, subject in enumerate(subjects):
|
| 68 |
+
dimensions = subject.get("dimensions", {})
|
| 69 |
+
frames = subject.get("frames", [])
|
| 70 |
+
label = subject.get('objectClass', f"Subject_{subject_idx}")
|
| 71 |
|
| 72 |
+
for frame_idx, frame in enumerate(frames):
|
| 73 |
+
rr.set_time_sequence("frame_idx", frame_idx)
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 74 |
|
| 75 |
+
try:
|
| 76 |
+
box_params = create_subject_box(frame, dimensions)
|
| 77 |
+
center = box_params['center']
|
| 78 |
+
half_size = box_params['half_size']
|
| 79 |
+
|
| 80 |
+
# Convert Euler angles to quaternion for rotation
|
| 81 |
+
rotation_q = euler_to_quaternion(frame['rotation'])
|
| 82 |
+
|
| 83 |
+
# 1) Log a transform for the subject so the box is placed at 'center' with correct rotation
|
| 84 |
+
rr.log(
|
| 85 |
+
f"world/subjects/subject_{subject_idx}",
|
| 86 |
+
rr.Transform3D(
|
| 87 |
+
translation=center,
|
| 88 |
+
rotation=rr.Quaternion(xyzw=rotation_q)
|
| 89 |
+
)
|
| 90 |
+
)
|
| 91 |
+
|
| 92 |
+
# 2) Log a box at the origin of the subject's local transform
|
| 93 |
+
rr.log(
|
| 94 |
+
f"world/subjects/subject_{subject_idx}/box",
|
| 95 |
+
rr.Boxes3D(
|
| 96 |
+
centers=np.array([[0.0, 0.0, 0.0]], dtype=np.float32),
|
| 97 |
+
half_sizes=np.array([half_size], dtype=np.float32),
|
| 98 |
+
colors=np.array([[0.8, 0.2, 0.2, 1.0]], dtype=np.float32),
|
| 99 |
+
labels=[label],
|
| 100 |
+
show_labels=False,
|
| 101 |
+
fill_mode="solid"
|
| 102 |
+
)
|
| 103 |
+
)
|
| 104 |
+
except Exception as e:
|
| 105 |
+
print(f"Error creating box parameters for subject {subject_idx}, frame {frame_idx}: {str(e)}")
|
| 106 |
|
| 107 |
def log_camera_trajectory(self, camera_frames: List[Dict[str, Any]]) -> None:
|
| 108 |
if not camera_frames:
|
|
|
|
| 116 |
"world/camera_trajectory",
|
| 117 |
rr.Points3D(
|
| 118 |
camera_positions,
|
| 119 |
+
colors=np.full((len(camera_positions), 4), [0.0, 0.8, 0.8, 1.0])
|
| 120 |
+
),
|
| 121 |
+
timeless=True
|
| 122 |
)
|
| 123 |
|
| 124 |
if len(camera_positions) > 1:
|
| 125 |
+
lines = np.stack([camera_positions[:-1], camera_positions[1:]], axis=1)
|
|
|
|
| 126 |
rr.log(
|
| 127 |
"world/camera_trajectory/line",
|
| 128 |
rr.LineStrips3D(
|
| 129 |
lines,
|
| 130 |
colors=[(0.0, 0.8, 0.8, 1.0)]
|
| 131 |
+
),
|
| 132 |
+
timeless=True
|
| 133 |
)
|
| 134 |
|
| 135 |
except Exception as e:
|
| 136 |
print(f"Error logging camera trajectory: {str(e)}")
|
| 137 |
|
| 138 |
+
def log_camera_frames(self, camera_frames: List[Dict[str, Any]]) -> None:
|
|
|
|
|
|
|
|
|
|
| 139 |
if not camera_frames:
|
| 140 |
return
|
| 141 |
|
|
|
|
|
|
|
|
|
|
|
|
|
| 142 |
for frame_idx, camera_frame in enumerate(camera_frames):
|
| 143 |
try:
|
| 144 |
rr.set_time_sequence("frame_idx", frame_idx)
|
|
|
|
| 163 |
)
|
| 164 |
)
|
| 165 |
|
|
|
|
|
|
|
|
|
|
| 166 |
except Exception as e:
|
| 167 |
print(f"Error logging camera frame {frame_idx}: {str(e)}")
|
| 168 |
|
|
|
|
| 179 |
helper_positions,
|
| 180 |
radii=np.full(len(helper_positions), 0.03),
|
| 181 |
colors=np.full((len(helper_positions), 4), [1.0, 1.0, 0.0, 1.0]),
|
| 182 |
+
),
|
| 183 |
+
timeless=True
|
| 184 |
)
|
| 185 |
|
| 186 |
for keyframe_idx, helper_keyframe in enumerate(helper_keyframes):
|
| 187 |
try:
|
| 188 |
+
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|