Temporally Consistent Long-Term Memory for 3D Single Object Tracking

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Yoo, Jaejoon, Lee, SuBeen, Jeon, Yerim, Lee, Miso, Heo, Jae-Pil
Format: Preprint
Published: 2026
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866908966394527744
author Yoo, Jaejoon
Lee, SuBeen
Jeon, Yerim
Lee, Miso
Heo, Jae-Pil
author_facet Yoo, Jaejoon
Lee, SuBeen
Jeon, Yerim
Lee, Miso
Heo, Jae-Pil
contents 3D Single Object Tracking (3D-SOT) aims to localize a target object across a sequence of LiDAR point clouds, given its 3D bounding box in the first frame. Recent methods have adopted a memory-based approach to utilize previously observed features of the target object, but remain limited to only a few recent frames. This work reveals that their temporal capacity is fundamentally constrained to short-term context due to severe temporal feature inconsistency and excessive memory overhead. To this end, we propose a robust long-term 3D-SOT framework, ChronoTrack, which preserves the temporal feature consistency while efficiently aggregating the diverse target features via long-term memory. Based on a compact set of learnable memory tokens, ChronoTrack leverages long-term information through two complementary objectives: a temporal consistency loss and a memory cycle consistency loss. The former enforces feature alignment across frames, alleviating temporal drift and improving the reliability of proposed long-term memory. In parallel, the latter encourages each token to encode diverse and discriminative target representations observed throughout the sequence via memory-point-memory cyclic walks. As a result, ChronoTrack achieves new state-of-the-art performance on multiple 3D-SOT benchmarks, demonstrating its effectiveness in long-term target modeling with compact memory while running at real-time speed of 42 FPS on a single RTX 4090 GPU. The code is available at https://github.com/ujaejoon/ChronoTrack
format Preprint
id arxiv_https___arxiv_org_abs_2604_13789
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Temporally Consistent Long-Term Memory for 3D Single Object Tracking
Yoo, Jaejoon
Lee, SuBeen
Jeon, Yerim
Lee, Miso
Heo, Jae-Pil
Computer Vision and Pattern Recognition
3D Single Object Tracking (3D-SOT) aims to localize a target object across a sequence of LiDAR point clouds, given its 3D bounding box in the first frame. Recent methods have adopted a memory-based approach to utilize previously observed features of the target object, but remain limited to only a few recent frames. This work reveals that their temporal capacity is fundamentally constrained to short-term context due to severe temporal feature inconsistency and excessive memory overhead. To this end, we propose a robust long-term 3D-SOT framework, ChronoTrack, which preserves the temporal feature consistency while efficiently aggregating the diverse target features via long-term memory. Based on a compact set of learnable memory tokens, ChronoTrack leverages long-term information through two complementary objectives: a temporal consistency loss and a memory cycle consistency loss. The former enforces feature alignment across frames, alleviating temporal drift and improving the reliability of proposed long-term memory. In parallel, the latter encourages each token to encode diverse and discriminative target representations observed throughout the sequence via memory-point-memory cyclic walks. As a result, ChronoTrack achieves new state-of-the-art performance on multiple 3D-SOT benchmarks, demonstrating its effectiveness in long-term target modeling with compact memory while running at real-time speed of 42 FPS on a single RTX 4090 GPU. The code is available at https://github.com/ujaejoon/ChronoTrack
title Temporally Consistent Long-Term Memory for 3D Single Object Tracking
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2604.13789