I have installed Caffe2 (using Anaconda package for python 2.7) and Detectron in an Ubuntu 18.04 system. Both installations were successful without errors and I have verified them.
However, on running the given detectron example infer_simple with the command given here,
python2 tools/infer_simple.py \
--cfg configs/12_2017_baselines/e2e_mask_rcnn_R-101-FPN_2x.yaml \
--output-dir /tmp/detectron-visualizations \
--image-ext jpg \
--wts https://s3-us-west-2.amazonaws.com/detectron/35861858/12_2017_baselines/e2e_mask_rcnn_R-101-FPN_2x.yaml.02_32_51.SgT4y1cO/output/train/coco_2014_train:coco_2014_valminusminival/generalized_rcnn/model_final.pkl \
demo
the script is executed without outputting the expected pdf images. The terminal output and nvidia-smi shows that inference is being carried out.
On using the argument --output always which outputs pdf images even when there aren't any detections I am able to produce the output pdfs (however indicating no detection boxes).
I have tried reducing the threshold value for detections with no luck.
Why is detectron inference not detecting (and outputting) any objects?
