# Serial Over Parallel: Learning Continual Unification for Multi-Modal Visual Object Tracking and Benchmarking

arXiv: https://arxiv.org/abs/2508.10655 · HTML: https://mlsanity.com/paper/2508.10655/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 20 of its 44 references that are in the map.
- First posted on arXiv 14 Aug 2025.

## Builds on (most cited first, 20 of 20)

- [TrackingNet: A Large-Scale Dataset and Benchmark for Object Tracking in the Wild](https://mlsanity.com/paper/1803.10794/) (2018)
- [Joint Feature Learning and Relation Modeling for Tracking: A One-Stream Framework](https://mlsanity.com/paper/2203.11991/) (2022)
- [VisEvent: Reliable Object Tracking via Collaboration of Frame and Event Flows](https://mlsanity.com/paper/2108.05015/) (2021)
- [ODTrack: Online Dense Temporal Token Learning for Visual Tracking](https://mlsanity.com/paper/2401.01686/) (2024)
- [RGB-T Object Tracking:Benchmark and Baseline](https://mlsanity.com/paper/1805.08982/) (2018)
- [Visible-Thermal UAV Tracking: A Large-Scale Benchmark and New Baseline](https://mlsanity.com/paper/2204.04120/) (2022)
- [Object Tracking by Jointly Exploiting Frame and Event Domain](https://mlsanity.com/paper/2109.09052/) (2021)
- [DepthTrack : Unveiling the Power of RGBD Tracking](https://mlsanity.com/paper/2108.13962/) (2021)
- [Single-Model and Any-Modality for Video Object Tracking](https://mlsanity.com/paper/2311.15851/) (2023)
- [OneTracker: Unifying Visual Object Tracking with Foundation Models and Efficient Tuning](https://mlsanity.com/paper/2403.09634/) (2024)
- [Towards Unified Token Learning for Vision-Language Tracking](https://mlsanity.com/paper/2308.14103/) (2023)
- [Prompting for Multi-Modal Tracking](https://mlsanity.com/paper/2207.14571/) (2022)
- [Exploiting Multimodal Spatial-temporal Patterns for Video Object Tracking](https://mlsanity.com/paper/2412.15691/) (2024)
- [Bi-directional Adapter for Multi-modal Tracking](https://mlsanity.com/paper/2312.10611/) (2023)
- [RGBD1K: A Large-scale Dataset and Benchmark for RGB-D Object Tracking](https://mlsanity.com/paper/2208.09787/) (2022)
- [DCEvo: Discriminative Cross-Dimensional Evolutionary Learning for Infrared and Visible Image Fusion](https://mlsanity.com/paper/2503.17673/) (2025)
- [Generative-based Fusion Mechanism for Multi-Modal Tracking](https://mlsanity.com/paper/2309.01728/) (2023)
- [Exploring Fusion Strategies for Accurate RGBT Visual Object Tracking](https://mlsanity.com/paper/2201.08673/) (2022)
- [TENet: Targetness Entanglement Incorporating with Multi-Scale Pooling and Mutually-Guided Fusion for RGB-E Object Tracking](https://mlsanity.com/paper/2405.05004/) (2024)
- [RGB-T Tracking via Multi-Modal Mutual Prompt Learning](https://mlsanity.com/paper/2308.16386/) (2023)

## Abstract (arXiv)

Unifying multiple multi-modal visual object tracking (MMVOT) tasks draws increasing attention due to the complementary nature of different modalities in building robust tracking systems. Existing practices mix all data sensor types in a single training procedure, structuring a parallel paradigm from the data-centric perspective and aiming for a global optimum on the joint distribution of the involved tasks. However, the absence of a unified benchmark where all types of data coexist forces evaluations on separated benchmarks, causing textitinconsistency between training and testing, thus leading to performance textitdegradation. To address these issues, this work advances in two aspects: ding182 A unified benchmark, coined as UniBench300, is introduced to bridge the inconsistency by incorporating multiple task data, reducing inference passes from three to one and cutting time consumption by 27%. ding183 The unification process is reformulated in a serial format, progressively integrating new tasks. In this way, the performance degradation can be specified as knowledge forgetting of previous tasks, which naturally aligns with the philosophy of continual learning (CL), motivating further exploration of injecting CL into the unification process. Extensive experiments conducted on two baselines and four benchmarks demonstrate the significance of UniBench300 and the superiority of CL in supporting a stable unification process. Moreover, while conducting dedicated analyses, the performance degradation is found to be negatively correlated with network capacity. Additionally, modality discrepancies contribute to varying degradation levels across tasks (RGBT > RGBD > RGBE in MMVOT), offering valuable insights for future multi-modal vision research. Source codes and the proposed benchmark is available at textithttps://github.com/Zhangyong-Tang/UniBench300.