Very interesting work and congrats for being accepted by ICCV. I am curious that have you tried your thinet model as backbone and applied it to other tasks like object detection, in faster-rcnn or ssd or some other tasks, like segementation and what is the performance ?
Very interesting work and congrats for being accepted by ICCV. I am curious that have you tried your thinet model as backbone and applied it to other tasks like object detection, in faster-rcnn or ssd or some other tasks, like segementation and what is the performance ?