CLIP-MoE: Mixture of Experts for CLIP

🚀 ClipServe: A fast API server for embedding text, images, and performing zero-shot classification using OpenAI’s CLIP model. Powered by FastAPI, Redis, and CUDA for lightning-fast, scalable AI applications. Transform texts and images into embeddings or classify images with custom labels—all through easy-to-use endpoints. 🌐📊

A novel naturalisitc real-world reinforement learning (RL) paradigm in Google street-view environments.

Zero-shot object detection with CLIP, utilizing Faster R-CNN for region proposals.

Zero-shot image classification performed using OpenAI's CLIP. Used the huggingface version of the CLIP model.

Using OpenAI's CLIP for Zero-Shot Image Search on the ImageNetFE dataset. Implemented using the huggingface framework.

Facial Expression Recognition using CLIP encoder as Feature Extractor

Deep learning pet breed recognition app

Fine-tune OpenAI's CLIP model for classification tasks

CLIP (Contrastive Language–Image Pre-training) for Bangla.

Search for images using text and images.

Computationally-free personalization at test time for sEMG gesture classification. Fast (gpu/cpu) ninapro API.

Visual semantic search system | Search across products | Text Query --> Visual Retrieval

Text2ImageDescription retrieves relevant images from Pascal VOC 2012 dataset using OpenAI CLIP, based on text queries, and generates descriptions using quantized Mistral-7b model.

Generation of faces, numbers and images...And Stable-Diffusion Inpainting through Segmentation through SAM and CLIP Model

Official implementation for "Blended Diffusion for Text-driven Editing of Natural Images" [CVPR 2022]

An experiment with movie scenes and contrastive learning

Simple implementation of OpenAI CLIP model in PyTorch.

Object tracking implemented with the Roboflow Inference API, DeepSort, and OpenAI CLIP.

Experiments with CLIP