FCN-style semantic segmentation implementation with mIoU evaluation and a custom segmentation model.

Profile README focused on AI implementation proof and deep learning paper implementations.

Spatial Transformer Network implementation with affine grid sampling, visualization, and baseline comparison.

PyTorch implementation proof for VGG and ResNet architectures on CIFAR-10.

CAM and Grad-CAM implementation proof with gradient hooks, heatmaps, and guided backpropagation.

PyTorch implementation proof: manual training loops for linear/logistic regression and feedforward MNIST classifiers.

FSRCNN super-resolution implementation with patch datasets, PSNR evaluation, and bicubic comparison.

RetinaNet object detection implementation with FPN, detection heads, and focal loss.