Scripts and data accompanying manuscript about flower cluster regression.
The regression models can be run via script multivew/run_experiments.py
.
Switching models and other experiment properties is done by arguments in function main
,
namely:
ground_truth = True / False
switch between field-validated (True) and visually-estimated (False) values, and
model_selector = "FPN" / "without_FPN" / "CountNet"
switch between different models.
The YOLO based experiment can be run via script yolo/run_experiments.py
.
In folder paper
are additional helper scripts for results aggregation etc.