I don't understand why you use other's Faster-rcnn, "Setup Faster R-CNN using instructions given in https://github.com/jwyang/faster-rcnn.pytorch". And the content in this new link is confusing. What does our RPN of SketchGuidedLocalization actually need?
And there are also some problems about torch version.
I didn't finish these steps in the introduction section and run the code directly.
I don't understand why you use other's Faster-rcnn, "Setup Faster R-CNN using instructions given in https://github.com/jwyang/faster-rcnn.pytorch". And the content in this new link is confusing. What does our RPN of SketchGuidedLocalization actually need? And there are also some problems about torch version.
I didn't finish these steps in the introduction section and run the code directly.