Pixie8888

Thanks for the quick reply! Does that mean: In opencv function, to compute backward optical flow, just simply change prvs to img_t+1, next to img_t? prvs = img_t+1 next =...

Hi, I use the evaluation code below: for skimage.measure.compare_ssim, I calculate ssim along each channel, then average them.  gt[t][i].shape = (b,3,h,w), value range(0,1), numpy float16 result:(0.9767,0.9038,0.8849) for pytorch_msssim: from...

Hi, I didn't solve these problems. Neither did I reproduce results on S3DIS.

@Cindy0725 Did you successfully run imvoxelnet on ScanNet using pytorch1.13+cu11.6? I am facing similar issue when using higher version of pytorch.

Hi realningzheng, I encountered similar situation when doing evaluation. I also found that inside 'solotest.json', the frame number is not consistent with the processed data. Did you solve the problem?

> Hi, sorry for some mass in test code. The evaluation procedure for stage one, single source localization, should be > > 1. python3 training_stage_one.py --mode test --ckpt_file path/to/ckpt >...

Hi @machadogj , How large is the 2,077 mp4 videos? I downloaded similar number of videos, but they only takes up around 40 GB. Is it normal?

Hi, Why not using SVD to get the left/right singular vectors as the basis?

Thanks! By the way, I am running objectsdf++ on scannet scene 1 on one 3090Ti, but the training is very slow, probably 72h to finish 2000 epochs. Is it normal?

oh Thanks! Do you have per scene result on ScanNet?