Spaces:
Sleeping
Sleeping
File size: 9,838 Bytes
d725335 20905e1 d725335 114ea19 8536e24 114ea19 8536e24 d725335 20905e1 d725335 20905e1 114ea19 8536e24 114ea19 8536e24 114ea19 8536e24 20905e1 d725335 20905e1 d725335 20905e1 d725335 20905e1 d725335 | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 | from __future__ import annotations
from pathlib import Path
from typing import Any
from tiny_trigger.models import ActionEvent, Detection
from tiny_trigger.video import _create_browser_mp4_writer, process_video, render_automation_video
class FakeDetector:
class_names = ["cat"]
def detect(
self,
frame: Any,
*,
frame_index: int,
timestamp_sec: float,
confidence: float,
image_size: int | None = None,
max_detections: int | None = None,
) -> list[Detection]:
assert image_size == 960
assert max_detections == 20
return [
Detection(
frame_index=frame_index,
timestamp_sec=timestamp_sec,
label="cat",
confidence=0.99,
bbox_xyxy=(2.0, 2.0, 12.0, 12.0),
bbox_xyxy_norm=(0.1, 0.1, 0.6, 0.6),
)
]
class TrackedDetector:
class_names = ["cat"]
def detect(
self,
frame: Any,
*,
frame_index: int,
timestamp_sec: float,
confidence: float,
image_size: int | None = None,
max_detections: int | None = None,
) -> list[Detection]:
offset = frame_index * 0.01
return [
Detection(
frame_index=frame_index,
timestamp_sec=timestamp_sec,
label="cat",
confidence=0.99,
bbox_xyxy=(2.0 + frame_index, 2.0, 12.0 + frame_index, 12.0),
bbox_xyxy_norm=(0.1 + offset, 0.1, 0.2 + offset, 0.2),
track_id=7,
)
]
class DuplicateDetector:
class_names = ["cat"]
def detect(
self,
frame: Any,
*,
frame_index: int,
timestamp_sec: float,
confidence: float,
image_size: int | None = None,
max_detections: int | None = None,
) -> list[Detection]:
return [
Detection(
frame_index=frame_index,
timestamp_sec=timestamp_sec,
label="cat",
confidence=0.62,
bbox_xyxy=(2.0, 2.0, 16.0, 16.0),
bbox_xyxy_norm=(0.1, 0.1, 0.5, 0.5),
),
Detection(
frame_index=frame_index,
timestamp_sec=timestamp_sec,
label="cat",
confidence=0.91,
bbox_xyxy=(3.0, 3.0, 17.0, 17.0),
bbox_xyxy_norm=(0.11, 0.11, 0.51, 0.51),
),
]
def test_process_video_with_fake_detector(tmp_path: Path) -> None:
cv2 = __import__("cv2")
video_path = _make_video(tmp_path)
result = process_video(
video_path=str(video_path),
class_prompt="cat",
frame_stride=2,
max_frames=2,
image_size=960,
max_detections=20,
detector=FakeDetector(),
output_dir=str(tmp_path),
)
assert Path(result.output_video_path).exists()
assert result.processed_frames == 2
assert [item.frame_index for item in result.detections] == [0, 2]
assert result.frame_stride == 2
assert result.sample_interval_sec is None
capture = cv2.VideoCapture(result.output_video_path)
try:
assert capture.get(cv2.CAP_PROP_FRAME_COUNT) == 4
assert capture.get(cv2.CAP_PROP_FPS) == 10.0
finally:
capture.release()
def test_process_video_does_not_synthesize_track_ids(tmp_path: Path) -> None:
video_path = _make_video(tmp_path, fps=10.0, frames=4)
result = process_video(
video_path=str(video_path),
class_prompt="cat",
frame_stride=1,
max_frames=3,
detector=FakeDetector(),
image_size=960,
max_detections=20,
output_dir=str(tmp_path),
)
assert [item.track_id for item in result.detections] == [None, None, None]
def test_process_video_preserves_detector_track_ids(tmp_path: Path) -> None:
video_path = _make_video(tmp_path, fps=10.0, frames=4)
result = process_video(
video_path=str(video_path),
class_prompt="cat",
frame_stride=1,
max_frames=3,
detector=TrackedDetector(),
output_dir=str(tmp_path),
)
assert [item.track_id for item in result.detections] == [7, 7, 7]
def test_process_video_suppresses_duplicate_same_label_boxes(tmp_path: Path) -> None:
video_path = _make_video(tmp_path, fps=10.0, frames=2)
result = process_video(
video_path=str(video_path),
class_prompt="cat",
frame_stride=1,
max_frames=1,
detector=DuplicateDetector(),
output_dir=str(tmp_path),
)
assert len(result.detections) == 1
assert result.detections[0].confidence == 0.91
def test_process_video_samples_once_per_second(tmp_path: Path) -> None:
video_path = _make_video(tmp_path, fps=30.0, frames=95)
result = process_video(
video_path=str(video_path),
class_prompt="cat",
frame_stride=2,
sample_interval_sec=1.0,
max_frames=3,
image_size=960,
max_detections=20,
detector=FakeDetector(),
output_dir=str(tmp_path),
)
assert result.frame_stride == 30
assert result.sample_interval_sec == 1.0
assert [item.frame_index for item in result.detections] == [0, 30, 60]
def test_process_video_samples_half_second_intervals(tmp_path: Path) -> None:
video_path = _make_video(tmp_path, fps=10.0, frames=16)
result = process_video(
video_path=str(video_path),
class_prompt="cat",
sample_interval_sec=0.5,
max_frames=3,
image_size=960,
max_detections=20,
detector=FakeDetector(),
output_dir=str(tmp_path),
)
assert result.frame_stride == 5
assert [item.frame_index for item in result.detections] == [0, 5, 10]
def test_render_automation_video_with_fired_event(tmp_path: Path) -> None:
video_path = _make_video(tmp_path)
detections = [
Detection(
frame_index=0,
timestamp_sec=0.0,
label="cat",
confidence=0.99,
bbox_xyxy=(2.0, 2.0, 12.0, 12.0),
bbox_xyxy_norm=(0.1, 0.1, 0.6, 0.6),
)
]
events = [
ActionEvent(
rule="cat-rule",
action="feed cat",
type="simulate",
frame_index=0,
timestamp_sec=0.0,
status="simulated",
)
]
output_path = render_automation_video(
source_video_path=str(video_path),
detections=detections,
events=events,
frame_stride=2,
max_frames=2,
output_dir=str(tmp_path),
)
assert Path(output_path).exists()
assert output_path.endswith("-automated.mp4")
def test_render_automation_video_uses_computed_stride(tmp_path: Path) -> None:
video_path = _make_video(tmp_path, fps=30.0, frames=95)
result = process_video(
video_path=str(video_path),
class_prompt="cat",
sample_interval_sec=1.0,
max_frames=3,
image_size=960,
max_detections=20,
detector=FakeDetector(),
output_dir=str(tmp_path),
)
output_path = render_automation_video(
source_video_path=str(video_path),
detections=result.detections,
events=[],
frame_stride=result.frame_stride,
max_frames=3,
output_dir=str(tmp_path),
)
assert Path(output_path).exists()
def test_process_video_writes_full_motion_clip_with_sampled_overlays(tmp_path: Path) -> None:
cv2 = __import__("cv2")
video_path = _make_video(tmp_path, fps=5.0, frames=15)
result = process_video(
video_path=str(video_path),
class_prompt="cat",
frame_stride=5,
max_frames=3,
image_size=960,
max_detections=20,
detector=FakeDetector(),
output_dir=str(tmp_path),
)
capture = cv2.VideoCapture(result.output_video_path)
try:
assert capture.get(cv2.CAP_PROP_FPS) == 5.0
assert capture.get(cv2.CAP_PROP_FRAME_COUNT) == 15
finally:
capture.release()
def test_process_video_writes_faststart_mp4(tmp_path: Path) -> None:
video_path = _make_video(tmp_path, fps=5.0, frames=6)
result = process_video(
video_path=str(video_path),
class_prompt="cat",
frame_stride=2,
max_frames=2,
image_size=960,
max_detections=20,
detector=FakeDetector(),
output_dir=str(tmp_path),
)
data = Path(result.output_video_path).read_bytes()
assert data.find(b"moov") < data.find(b"mdat")
def test_browser_mp4_writer_uses_mp4v_only(monkeypatch, tmp_path: Path) -> None:
calls: list[str] = []
class Writer:
def isOpened(self) -> bool:
return True
def release(self) -> None:
return None
class CV2:
@staticmethod
def VideoWriter_fourcc(*codec):
calls.append("".join(codec))
return 1234
@staticmethod
def VideoWriter(path, fourcc, fps, frame_size):
return Writer()
monkeypatch.setitem(__import__("sys").modules, "cv2", CV2)
writer = _create_browser_mp4_writer(tmp_path / "out.mp4", 8.0, (32, 32))
assert writer is not None
assert calls == ["mp4v"]
def _make_video(tmp_path: Path, *, fps: float = 10.0, frames: int = 4) -> Path:
cv2 = __import__("cv2")
video_path = tmp_path / "input.mp4"
writer = cv2.VideoWriter(str(video_path), cv2.VideoWriter_fourcc(*"mp4v"), fps, (32, 32))
for index in range(frames):
frame = __import__("numpy").zeros((32, 32, 3), dtype="uint8")
frame[:] = (index * 20) % 256
writer.write(frame)
writer.release()
return video_path
|