This repository is a curated list of papers, datasets, and resources for Described Object Detection (DOD), Open-Vocabulary/Open-World Object Detection (OVD), and Referring Expression Comprehension (REC). It serves researchers and practitioners in computer vision and natural language processing looking to advance object detection capabilities beyond predefined categories.

How It Works

The repository organizes a vast collection of academic literature, categorizing papers by their specific task (DOD, OVD, REC) and highlighting key methods, datasets, and related resources. It provides links to papers, code repositories, and datasets, enabling users to quickly access and evaluate relevant research. The inclusion of leaderboards and survey papers offers a structured overview of the field's progress.

Quick Start & Requirements

This is a curated list, not a runnable software package. Accessing the resources requires individual setup for each paper's code or dataset.

Highlighted Details

• Comprehensive coverage of Described Object Detection (DOD) methods, including a leaderboard for up-to-date approaches.
• Extensive lists of papers and datasets for Open-Vocabulary Object Detection (OVD), covering various techniques from CLIP-based methods to diffusion models.
• Detailed resources for Referring Expression Comprehension (REC) and Visual Grounding, including benchmarks and datasets.
• Links to related survey papers and other "awesome" lists for broader context.

Maintenance & Community

The repository is actively maintained, with frequent updates and an open invitation for pull requests to include missing work. The structure is inspired by other popular "awesome" lists in the multimodal AI community.

Licensing & Compatibility

This repository itself is a list and does not have a specific license. Individual papers and code repositories linked within will have their own licenses.

Limitations & Caveats

As a curated list, it does not provide direct functionality or code. Users must independently find, download, and set up the resources mentioned. Some older "awesome" lists linked may not be actively maintained.