REMIND: RE-Identification with Memory for INDoor Navigation

2026-07-10

AI Summary

This work addresses the challenge of stable multi-object re-identification for indoor mobile robots under drastic viewpoint shifts and varying illumination during long-term operation. The authors propose an online re-identification method that operates without camera pose or depth information, innovatively integrating a dual-bank multi-prototype appearance memory, part-and-background-level descriptors, and a neighborhood context reasoning module to emulate human visual mechanisms of familiarity and spatial cues. Leveraging frozen DINOv3 features and Hungarian algorithm-based joint matching, the approach further incorporates an ambiguity-aware safeguarding strategy. Evaluated on a newly collected indoor revisiting dataset, the method achieves an IDF1 score of 90.35%, substantially outperforming existing video segmentation baselines, and demonstrates superior performance across most ScanNet++ scenes with improved memory efficiency.

Abstract

Mobile robots operating indoors must re-identify previously observed objects after long temporal gaps, significant viewpoint changes, and severe illumination variations. This remains a challenging problem: multi-object tracking methods are optimized for short-term association of pedestrians and vehicles at video rates, person and vehicle re-identification approaches lack persistent memory mechanisms, and state-of-the-art video object segmentation techniques rely on reactive distractor filtering rather than enforcing global identity consistency. To address these limitations, we present REMIND, an online tracker designed for long-term multi-object re-identification of generic indoor objects from monocular RGB imagery, requiring neither camera pose nor depth. Motivated by evidence from visual cognition that humans rely on accumulated appearance familiarity and spatial context rather than explicit self-localization, REMIND combines frozen DINOv3 features with a dual-bank multi-prototype appearance memory, part- and background-level descriptors, a neighbour-context reasoning module exploiting spatial co-occurrence, and joint Hungarian assignment with ambiguity-aware safeguards. On a purpose-built indoor dataset featuring controlled revisits and dense same-class clutter, REMIND reaches 90.35% IDF1, nearly 20 points above a state-of-the-art video object segmentation baseline and more than 36 above a strong tracking-by-detection baseline. On ScanNet++, it attains the highest IDF1 in every setting but one, end-to-end detection over all scenes, where the tracking-by-detection baseline is marginally ahead while REMIND still associates and recovers identities more accurately; it also completes every scene, whereas the video object segmentation baseline exhausts GPU memory on 66.9% under YOLO detections. The complete system, evaluation framework, and dataset are publicly released.

Problem

• re-identification
• indoor navigation
• long-term tracking
• multi-object association
• identity consistency

Innovation

• long-term re-identification
• appearance memory
• spatial context reasoning
• monocular RGB tracking
• identity consistency