Motion-Refined DINOSAUR for Unsupervised Multi-Object Discovery
- URL: http://arxiv.org/abs/2509.02545v1
- Date: Tue, 02 Sep 2025 17:44:49 GMT
- Title: Motion-Refined DINOSAUR for Unsupervised Multi-Object Discovery
- Abstract summary: Unsupervised multi-object discovery (MOD) aims to detect and localize distinct object instances in visual scenes without any form of human supervision.<n>Recent approaches leverage object-centric learning (OCL) and motion cues from video to identify individual objects.<n>We address this limitation with MR-DINOSAUR, a minimalistic unsupervised approach that extends the self-supervised pre-trained OCL model, DINOSAUR, to the task of unsupervised multi-object discovery.
- Score: 52.60873386395476
- License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
- Abstract: Unsupervised multi-object discovery (MOD) aims to detect and localize distinct object instances in visual scenes without any form of human supervision. Recent approaches leverage object-centric learning (OCL) and motion cues from video to identify individual objects. However, these approaches use supervision to generate pseudo labels to train the OCL model. We address this limitation with MR-DINOSAUR -- Motion-Refined DINOSAUR -- a minimalistic unsupervised approach that extends the self-supervised pre-trained OCL model, DINOSAUR, to the task of unsupervised multi-object discovery. We generate high-quality unsupervised pseudo labels by retrieving video frames without camera motion for which we perform motion segmentation of unsupervised optical flow. We refine DINOSAUR's slot representations using these pseudo labels and train a slot deactivation module to assign slots to foreground and background. Despite its conceptual simplicity, MR-DINOSAUR achieves strong multi-object discovery results on the TRI-PD and KITTI datasets, outperforming the previous state of the art despite being fully unsupervised.
Related papers
- Autonomous UAV Navigation for Individual Wildlife Re-Identification [0.350327094468706]
We propose an autonomous drone navigation system that actively optimize image capture for downstream re-ID.<n>The system combines YOLOv11 object detection with a DINOv2-based pose to guide real-time flight decisions.<n>We show the approach generalizes to other species with diagnostic surface patterns, including giraffes, tigers, and elephants.
arXiv Detail & Related papers (2026-06-30T14:56:04Z) - Lightweight Multi-Frame Integration for Robust YOLO Object Detection in Videos [11.532574301455854]
We propose a highly effective strategy for multi-frame video object detection.<n>Our method improves robustness, especially for lightweight models.<n>We contribute the BOAT360 benchmark dataset to support future research in multi-frame video object detection in challenging real-world scenarios.
arXiv Detail & Related papers (2025-06-25T15:49:07Z) - C-Drag: Chain-of-Thought Driven Motion Controller for Video Generation [81.4106601222722]
Trajectory-based motion control has emerged as an intuitive and efficient approach for controllable video generation.<n>We propose a Chain-of-Thought-based motion controller for controllable video generation, named C-Drag.<n>Our method includes an object perception module and a Chain-of-Thought-based motion reasoning module.
arXiv Detail & Related papers (2025-02-27T08:21:03Z) - Seamless Detection: Unifying Salient Object Detection and Camouflaged Object Detection [73.85890512959861]
We propose a task-agnostic framework to unify Salient Object Detection (SOD) and Camouflaged Object Detection (COD)<n>We design a simple yet effective contextual decoder involving the interval-layer and global context, which achieves an inference speed of 67 fps.<n> Experiments on public SOD and COD datasets demonstrate the superiority of our proposed framework in both supervised and unsupervised settings.
arXiv Detail & Related papers (2024-12-22T03:25:43Z) - Un-EVIMO: Unsupervised Event-Based Independent Motion Segmentation [29.557103057014988]
Event cameras are a novel type of biologically inspired vision sensor known for their high temporal resolution, high dynamic range, and low power consumption.<n>We propose the first event framework that generates IMO pseudo-labels using geometric constraints.<n>Due to its unsupervised nature, our method can handle an arbitrary number of not predetermined objects and is easily scalable to datasets where expensive IMO labels are not readily available.
arXiv Detail & Related papers (2023-11-30T18:59:32Z) - Object-Centric Multiple Object Tracking [124.30650395969126]
This paper proposes a video object-centric model for multiple-object tracking pipelines.
It consists of an index-merge module that adapts the object-centric slots into detection outputs and an object memory module.
Benefited from object-centric learning, we only require sparse detection labels for object localization and feature binding.
arXiv Detail & Related papers (2023-09-01T03:34:12Z) - Delving into Motion-Aware Matching for Monocular 3D Object Tracking [81.68608983602581]
We find that the motion cue of objects along different time frames is critical in 3D multi-object tracking.
We propose MoMA-M3T, a framework that mainly consists of three motion-aware components.
We conduct extensive experiments on the nuScenes and KITTI datasets to demonstrate our MoMA-M3T achieves competitive performance against state-of-the-art methods.
arXiv Detail & Related papers (2023-08-22T17:53:58Z) - Bridging the Gap to Real-World Object-Centric Learning [66.55867830853803]
We show that reconstructing features from models trained in a self-supervised manner is a sufficient training signal for object-centric representations to arise in a fully unsupervised way.
Our approach, DINOSAUR, significantly out-performs existing object-centric learning models on simulated data.
arXiv Detail & Related papers (2022-09-29T15:24:47Z)
This list is automatically generated from the titles and abstracts of the papers in this site.
This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.