ChidanandKumarKS

encoder , decoder framework for image reconstruction using event cameras

2

i tried to use stereospike which is using your repo to do image reconstruction from event camera rather than optical flow. But iam unable to get reconstruct image using event...

Dataset download issue?

2

Hi Sir, Unable to download dataset in INDIA, baidu websites are blocked. Requesting you to share the data in Google drive/One drive. Kindly do the needful

Hi, Ncaltech101 dataset downloaded from prophese contains .npy files whereas code takes *.bin files. Is Readme.txt is correct?

@Shathe Kindly requesting you to share the corresponding event data apart from 6-channel event image representation. Kindly do the needful

How many GPUs were required to replicate the same result?

4

Thanks for releasing the code. Kindly let me know How many GPUs were required to replicate the same result? Is the code is completed for training and inference? Any pretrained...

kindly share pretrained mode

where to get REsNet34 and ResNet50 pretrained models on ImageNet using InPlaceBN for CCD? Kindly help

Fake code released: Don't share wrong codes which will guide falsely

2

@RodrigoGantier I believe the code you have shared for training is not exactly reported in the paper. It seems you have hidden the codes deliberately or you are unable to...

After training, results are not the same as wrt pretrianed model?

1

@RodrigoGantier I tried to train SPADE-E2VID model with smaple dataset shared. The results are far from the pretrained model in terms of MSE/SSIM/LPIPS/TC. Kindly do the needful by giving few...