Refactor get_min_region_size for dimension normalization

Refactor get_min_region_size to normalize dimensions for smaller models and ensure minimum region size is 320 for larger models.
This commit is contained in:
Nicolas Mowen 2026-05-15 07:44:18 -06:00 committed by GitHub
parent 19c61f561f
commit 3c71816fd0
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194

View File

@ -271,18 +271,17 @@ def get_min_region_size(model_config: ModelConfig) -> int:
"""Get the min region size."""
largest_dimension = max(model_config.height, model_config.width)
if largest_dimension > 320:
# We originally tested allowing any model to have a region down to half of the model size
# but this led to many false positives. In this case we specifically target larger models
# which can benefit from a smaller region in some cases to detect smaller objects.
half = int(largest_dimension / 2)
# return largest dimension for smaller models, but make sure the dimension is normalized
if largest_dimension < 320:
if largest_dimension % 4 == 0:
return largest_dimension
if half % 4 == 0:
return half
return int((largest_dimension + 3) / 4) * 4
return int((half + 3) / 4) * 4
return largest_dimension
# Any model that is 320 or larger should have a minimum region size of 320
# this allows larger models to use smaller regions to detect smaller objects
# in the case that the motion area is smaller so that it can be upscaled.
return 320
def create_tensor_input(frame, model_config: ModelConfig, region):