All Projects → IbrahimSobh → Imageclassification

IbrahimSobh / Imageclassification

Licence: mit
Deep Learning: Image classification, feature visualization and transfer learning with Keras

Projects that are alternatives of or similar to Imageclassification

Keras transfer cifar10
Object classification with CIFAR-10 using transfer learning
Stars: ✭ 120 (+44.58%)
Mutual labels:  jupyter-notebook, image-classification, transfer-learning
Image classifier
CNN image classifier implemented in Keras Notebook 🖼️.
Stars: ✭ 139 (+67.47%)
Mutual labels:  jupyter-notebook, image-classification, transfer-learning
Cvpr18 Inaturalist Transfer
Large Scale Fine-Grained Categorization and Domain-Specific Transfer Learning. CVPR 2018
Stars: ✭ 164 (+97.59%)
Mutual labels:  jupyter-notebook, image-classification, transfer-learning
Graphrole
Automatic feature extraction and node role assignment for transfer learning on graphs (ReFeX & RolX)
Stars: ✭ 38 (-54.22%)
Mutual labels:  transfer-learning, feature-extraction
Seismic Transfer Learning
Deep-learning seismic facies on state-of-the-art CNN architectures
Stars: ✭ 32 (-61.45%)
Mutual labels:  jupyter-notebook, transfer-learning
Densedepth
High Quality Monocular Depth Estimation via Transfer Learning
Stars: ✭ 963 (+1060.24%)
Mutual labels:  jupyter-notebook, transfer-learning
Covid 19 Bert Researchpapers Semantic Search
BERT semantic search engine for searching literature research papers for coronavirus covid-19 in google colab
Stars: ✭ 23 (-72.29%)
Mutual labels:  jupyter-notebook, search-engine
Average Word2vec
🔤 Calculate average word embeddings (word2vec) from documents for transfer learning
Stars: ✭ 52 (-37.35%)
Mutual labels:  jupyter-notebook, transfer-learning
Computervision Recipes
Best Practices, code samples, and documentation for Computer Vision.
Stars: ✭ 8,214 (+9796.39%)
Mutual labels:  jupyter-notebook, image-classification
Mish
Official Repsoitory for "Mish: A Self Regularized Non-Monotonic Neural Activation Function" [BMVC 2020]
Stars: ✭ 1,072 (+1191.57%)
Mutual labels:  jupyter-notebook, image-classification
Weakly Supervised 3d Object Detection
Weakly Supervised 3D Object Detection from Point Clouds (VS3D), ACM MM 2020
Stars: ✭ 61 (-26.51%)
Mutual labels:  jupyter-notebook, transfer-learning
Deep learning projects
Stars: ✭ 28 (-66.27%)
Mutual labels:  jupyter-notebook, image-classification
Concise Ipython Notebooks For Deep Learning
Ipython Notebooks for solving problems like classification, segmentation, generation using latest Deep learning algorithms on different publicly available text and image data-sets.
Stars: ✭ 23 (-72.29%)
Mutual labels:  jupyter-notebook, image-classification
Teacher Student Training
This repository stores the files used for my summer internship's work on "teacher-student learning", an experimental method for training deep neural networks using a trained teacher model.
Stars: ✭ 34 (-59.04%)
Mutual labels:  jupyter-notebook, transfer-learning
Grocery Product Classification
Implementation of the paper "A Hierarchical Grocery Store Image Dataset with Visual and Semantic Labels"
Stars: ✭ 23 (-72.29%)
Mutual labels:  jupyter-notebook, image-classification
Alpha pooling
Code for our paper "Generalized Orderless Pooling Performs Implicit Salient Matching" published at ICCV 2017.
Stars: ✭ 51 (-38.55%)
Mutual labels:  jupyter-notebook, image-classification
Neural Painters X
Neural Paiters
Stars: ✭ 61 (-26.51%)
Mutual labels:  jupyter-notebook, transfer-learning
Sru Deeplearning Workshop
دوره 12 ساعته یادگیری عمیق با چارچوب Keras
Stars: ✭ 66 (-20.48%)
Mutual labels:  jupyter-notebook, transfer-learning
My Journey In The Data Science World
📢 Ready to learn or review your knowledge!
Stars: ✭ 1,175 (+1315.66%)
Mutual labels:  jupyter-notebook, feature-extraction
Tfidf
Simple TF IDF Library
Stars: ✭ 6 (-92.77%)
Mutual labels:  jupyter-notebook, feature-extraction

Practical Image Classifcation with Keras:

  • Hello Colab
  • Image Classification
  • Pretrained Feature Vizualization
  • Transfer Learning
  • Feature Based Image Search

01 HelloColab.ipynb

Basic Colab demo

colab

colab ex01 colab ex01


02 ImageClassification.ipynb

In this code, we are going to implement a basic image classifier:

  • Load the dataset (MNIST hand written digits)
  • Design a deep learning model and inspect its learnable parameters
  • Train the model on the training data and inspect learning curves
  • Evaluate the trained model on the never seen testing data
  • Save the model for later use
  • Load and use the model

MNIST class ex01


03 PretrainedFeatureViz .ipynb

In this code we are going to load pretrained image classification networks

  • ResNet50
  • VGG16
  • VGG19

deepnets

Then using a pretrained network, feature extraction and visualization is conducted via t-SNE

tsne


04 TransferLearning.ipynb

In Transfer learning, we would like to leverage the knowledge learned by a source task to help learning another target task. For example, a well-trained, rich image classification network could be leveraged for another image target related task. Another example, the knowledge learned by a network trained on simulated environment can be transferred to a network for the real environment. Basically, there are two basic scenarios for neural networks transfer learning:

  • Feature Extraction
  • Fine Tuning

A well known example for transfer learning is to load the already trained large scale classification VGG network that is able to classify images into one of 1000 classes, and use it for another task such as classification of special medical images.


05 FeatureBasedImageSearch.ipynb

Image search engines: Generally speaking, search engine usually takes a query and returns results. Image search engines takes an input image as an image query, then the image search engine finds the "similar" images within its indexed database and returns them as the search result. How to measure similarity between two images?

  • Pixel space: One simple way is to measure the euclidean distance between the two images in the pixel space. Accordingly, if two images have common or near values for the corresponding pixels, are then considered "similar". This could work some times, however a dolphin and air plane images with blue backgrounds will be considered similar from pixels point of view, and we do not want that!
  • Feature space: Another approach is to use the feature space instead of pixel space when computing the euclidean distance between the two images. In other words, project the images into a space where images with similar features are close to each others. In this space dolphins and airplanes are separated despite pixel level similarity.

search tsne search results

Note that the project description data, including the texts, logos, images, and/or trademarks, for each open source project belongs to its rightful owner. If you wish to add or remove any projects, please contact us at [email protected].