2

In this example you have a gray scale image of size 572x572 and 1 (gray) channel. The first convolution operation consists of 64 filters of size 3x3 and 1 channel per filter. The channel of the filters always fits the channel size of the previous layer (here: the Input). In the second convolution step of this explicit architecture, you again use 64 filters ...


1

The classification head works as follows. After the stack of BiFPN we have a feature map of size B x C x H x W. For EfficientDet H and W are 1/8 of the input image size. Then for each pixel in this feature map one applies one convolution to get the bounding boxes. The model predicts n_anchors - rescaled and shifted versions of reference boxes. The number of ...


1

Depth maps are created using principles of photometry (method of measuring light). The depth maps (rather images) you took from the website are "images" not exact depth "maps". So by default when you pull out a png image from a webpage, it will be saved in "RGB". That is the reason you got an array with 3 layers. In practice, it ...


1

Object detection models usually generate multiple detections per object. Duplicates are removed in a post-processing step called Non-Maximum Suppression (NMS). The Pytorch code that performs this post-processing is called here in the RegionProposalNetwork class. The filtering loop you've mentioned performs the NMS and applies the score_thresh threshold (...


1

What you want to do is called multi-task learning. Here's what you do: Create a second Input. Attach it to 1D CNN (2-3 layers), so it aggregates this tabular information. Concatenate this feature with the intermediate feature generated by the U-Net using Concatenate layer. Put a dense layer of 2 after this. Put softmax with units = number of classes. Add CE ...


1

Well, some time ago I also faced the same issue in the semantic segmentation task. Batch normalization is expected to improve convergence, because the normalization of activations prevents the explosion of the gradients magnitude and leads to more steady convergence. Adam is an adaptive optimizer with momentum and division by the weighted sum of gradients on ...


Only top voted, non community-wiki answers of a minimum length are eligible