ソンス

Results 46 issues of ソンス

hello author, for popular datasets and models, the metrics are calculated for validation set/ or both train an test? while writing journal papers. thank you

hello authors, please help me with some query. i see your unet model output is shape (batch, num_classes, h,w). i have similar unet model. but i want to know how...

hello authors, - can you please suggest me how to score the heatmap output compared wiht the ground truth labels. The iou or dice score similar to segmentation can be...

hello authors, great work from you guys. does this tool provide keypoints annotation? thank you

hello author, does this tensorflow implementation works for realtime webcam input? what fps did you get? thank you

hello author, I read your paper at https://arxiv.org/abs/1904.01941 I want to ask inorder to generate gaussian heatmaps, we need the center point coordinate information. Or the bounding box information of...

hello author, i believe that your goal is to predict individual charecters. When you generate the heatmap labels, you already use a model to predict individual charecters. How is the...

The image after transformation using the mean of imagenet gives darker image. Is this right? ![Figure_2](https://user-images.githubusercontent.com/30766648/167595955-5c933e41-a276-4642-b2b9-87929e160821.png) ![Figure_1](https://user-images.githubusercontent.com/30766648/167595973-22bcf0d9-c8e6-47dc-8f83-f4d332f3d3dd.png)

hello author, Did you use the the mirror and scale augmentaions in the dataloader. There are arguments in the dataloader function, but have not been used ? thank you

hello author, i see you have used only 3x3 and 7x7 filter size, any particular reason?