aditya1709

@Cartucho Thanks for the link. I used other literature to clear some doubts. But I am still confused - These were some of the results I got - 1. **mAP...

Apologies for the repeated question. Thanks for your patience. 1. After arranging them in descending order of confidences, we calculate precision and recall based on the cumulative count of True...

Have you figured it out if it has been updated in this repository ?

@MrWanter Can you please help me with what hyperparamter changes (tuning) you made to get the same result as the paper? My training process has been futile as even when...

@FrancescoFornasa - How many steps did it take for the network to converge and what was the batch size you used?

I tried out a few solutions, nothing seemed to work. It ended up training from the beginning. Have you made any progress on this front?

This might have happened if the .pth file was not downloaded properly. If you are using wget or curl try to check if it was downloaded correctly maybe by using...

Thanks for the response @Real-YeJ . Using the rectification network from ASTER with the cropped image from TextFuseNet output resulted in this ->  Any idea what might be wrong...

Thanks @asofter I have been working on this independently and would be down to collaborate if we can work on it together. I have a solution that I have almost...

I used the inbuilt pytorch profiler for a forward pass. All operations are consistently taking longer with FP16 ( apex ) than FP32. The quantity I compared was CUDA time...