DADO: A Depth-Attention framework for Object Discovery

Federico Gonzalez, Estefania Talavera, Petia Radeva
10/8/2025
cs.CV

No official code found. Best implementation: fedegonzal/dado (2 stars, MIT, last commit 2025-07-06).

Abstract

Unsupervised object discovery, the task of identifying and localizing objects in images without human-annotated labels, remains a significant challenge and a growing focus in computer vision. In this work, we introduce a novel model, DADO (Depth-Attention self-supervised technique for Discovering unseen Objects), which combines an attention mechanism and a depth model to identify potential objects in images. To address challenges such as noisy attention maps or complex scenes with varying depth planes, DADO employs dynamic weighting to adaptively emphasize attention or depth features based on the global characteristics of each image. We evaluated DADO on standard benchmarks, where it outperforms state-of-the-art methods in object discovery accuracy and robustness without the need for fine-tuning.

View on arXivView PDF

Code Implementations(1)

fedegonzal/dadoCommunity74%
MIT

DADO: A Depth-Attention framework for Object Discovery

20Jul 6, 20251 years agoMIT

Similar Papers

HEAP: Unsupervised Object Discovery and Localization with Contrastive Grouping

85% match2023

Weakly-supervised Contrastive Learning for Unsupervised Object Discovery

85% match1 repo62023

Unsupervised Object Discovery: A Comprehensive Survey and Unified Taxonomy

84% match2024

Local-Global Attention: An Adaptive Mechanism for Multi-Scale Feature Integration

83% match1 repo112024

Weakly Supervised Monocular 3D Detection with a Single-View Image

83% match2024

Prompt-Free Universal Region Proposal Network

83% match1 repo682026

Cite this paper

@article{gonzalez2025dado,
  title  = {DADO: A Depth-Attention framework for Object Discovery},
  author = {Federico Gonzalez and Estefania Talavera and Petia Radeva},
  year   = {2025},
  eprint = {2510.07089},
  archivePrefix = {arXiv},
  url    = {http://arxiv.org/abs/2510.07089v1}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/2510.07089.svg)](https://codeofpaper.com/paper/2510.07089)

Discussion