Hello, I'm very interested in your 2020 TTNet paper, I reproduced the code, but the result is very different from your demo, the ball detection error is very large, the semantic segmentation is not accurate, and there is no other detection data in your demo. Do you know what went wrong? Looking forward to hearing from you
Hello, I'm very interested in your 2020 TTNet paper, I reproduced the code, but the result is very different from your demo, the ball detection error is very large, the semantic segmentation is not accurate, and there is no other detection data in your demo. Do you know what went wrong? Looking forward to hearing from you