Which Faster R-CNN model is adopted to generate the object detetction results?

ASMIftekhar / VSGNet

VSGNet:Spatial Attention Network for Detecting Human Object Interactions Using Graph Convolutions.

MIT License

100 stars 20 forks source link

Which Faster R-CNN model is adopted to generate the object detetction results? #5

Closed MuchHair closed 4 years ago

MuchHair commented 4 years ago

Is the Faster R-CNN model also based on the ResNet-152 backbone that is used for feature exrtraction ？Thanks for reply.

oulutan commented 4 years ago

The object detector is separate and uses its own backbone. We used the tf object detection API with the architecture faster_rcnn_nas_coco_2018_01_28. We also provide the object detection results that we generated. If you use the bash download_data.sh, it will download the object detection results as well