kaggle Dog Breed Identification

"Exploring Vision Transformers for Fine-grained Classification" at CVPRW FGVC8

This project is a study that compares two machine learning approches (Feature extraction with SIFT descriptors and deep learning) in order to classify dogs races (Stanford dogs dataset). A flask application is built from the weights of the final deep neural network trained : http://bit.ly/mk_cv_dogs

The source code for Multi-Scale Kronecker-Product Relation Networks for Few-Shot Learning

Dogs Breeds Classification With TFLite Using Stanford Dogs Breeds Dataset.

Official implementation of the paper: Learn to aggregate global and local representations for few-shot learning

FGVC project with the Stanford Dogs dataset.

Transfer Learning | Image Classification | CNN | Jupyter Notebook | Python | TensorFlow | Stanford Dogs Dataset

ML + MobileApp = JaDIS

This repository performs Computer-Vision tasks on multiple Image Datasets using CNN based Networks.

🐕 От фотографии к породе за 50 миллисекунд • Нейросеть на базе MobileNetV2 распознаёт 120 пород собак с точностью 85-90% • Трансферное обучение, тонкая настройка, объяснимый ИИ • Применение: ветеринария, приюты, образование • TensorFlow/Keras

Stanford dogs dataset breed classification with Xception (CNN)

EfficientNetB7 applied to the Stanford Dogs dataset, achieving 92.76% training accuracy, 95.64% validation accuracy, and 95.64% test accuracy.

Using transfer learning on a CNN to build a 120 breed dog classifier

Tuning different Pytorch/Tensorflow pre-trained models like ResNet50 , Wide ResNet_50.2 , VGG16 and a custom CNN model to classify a dog image among 120 breeds

Applying Transfer Learning on Stanford Dogs Dataset

Utilizing CNNs for image classification of 120 dog breeds in the Stanford Dogs Dataset.