Personal AI search copilot, open-source Perplexity

Awesome-LLM-3D: a curated list of Multi-modal Large Language Model in 3D world Resources

Leveraging system development and robot deployment for ground-based autonomous navigation and exploration.

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

This is an official implementation for "Swin Transformer: Hierarchical Vision Transformer using Shifted Windows" on Object Detection and Instance Segmentation.

设计模式超简单的解释 - An ultra-simplified explanation to design patterns

Implementation / replication of DALL-E, OpenAI's Text to Image Transformer, in Pytorch

TensorFlow code and pre-trained models for BERT and ERNIE

PyTorch bottom-up attention with Detectron2

File Picker for React Native

RTI's open-source Call for Service (CFS) application for showing CFS/911 data.