Leveraging the extensive training data from SA-1B, the segment anything model (SAM) demonstrates remarkable generalization and zero-shot capabilities. However, as a category-agnostic instance ...

This repo implements UniTok, a unified visual tokenizer well-suited for both generation and understanding tasks. It is compatiable with autoregressive generative models (e.g. LlamaGen), multimodal ...

You're scrolling through your feed. What made you stop? Was it a wall of text… or was it something visual? Your prospects are making snap judgments about your business in milliseconds. And in today's ...

Abstract: Unmanned aerial vehicles (UAVs) have found numerous applications and are expected to bring fertile business opportunities in the next decade. Among various enabling technologies for UAVs, ...

3D Visual Grounding (3DVG) aims to locate objects in 3D scenes based on textual descriptions, which is essential for applications like augmented reality and robotics. Traditional 3DVG approaches rely ...