imagecode

Image retrieval benchmark

Provides code and data for an image retrieval benchmark that tests contextual understanding of text descriptions with images.

Code and data for ImageCoDe, a contextual vison-and-language benchmark

GitHub

39 stars
4 watching
6 forks
Language: Python
last commit: over 2 years ago

Related projects:

RepositoryDescriptionStars
kirk86/imageretrievalThis project provides tools and techniques for image retrieval based on visual features.56
cleanlab/cleanvisionAutomatically detects issues in image datasets to improve computer vision models1,038
jaiverma/matlabA collection of MATLAB code examples for various digital image processing tasks.37
filipradenovic/cnnimageretrievalThis MATLAB toolbox trains and tests convolutional neural networks for image retrieval tasks, including fine-tuning and supervised whitening.188
jcupitt/libvipsAn image processing library with low memory needs53
ibm/max-resnet-50An image classification model using the ResNet-50 architecture, trained on the ImageNet dataset.14
ibm/max-inception-resnet-v2An image classification model using a third-generation deep residual network.27
ailab-cvc/seed-benchA benchmark for evaluating large language models' ability to process multimodal input322
damo-nlp-sg/m3examA benchmark for evaluating large language models in multiple languages and formats93
szilard/benchm-mlA benchmark for evaluating machine learning algorithms' performance on large datasets1,874
isekai-portal/link-context-learningAn implementation of a multimodal learning approach to improve language models' ability to recognize unseen images and understand novel concepts.91
luispedro/mahotasA library of fast computer vision algorithms implemented in C++ for speed, operating over numpy arrays.855
mlcommons/inferenceMeasures the performance of deep learning models in various deployment scenarios.1,256
cambridge-mlg/fitThis repository provides code for a few-shot transfer learning approach to personalized and federated image classification11
aifeg/benchlmmAn open-source benchmarking framework for evaluating cross-style visual capability of large multimodal models84