AiATrack: Attention in Attention for Transformer Visual Tracking

Gao, Shenyuan; Zhou, Chunluan; Ma, Chao; Wang, Xinggang; Yuan, Junsong

doi:10.48550/arxiv.2207.09603

Cited by 3 publications

(3 citation statements)

References 61 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Trackers employing transformers can be categorized into two classes: CNN transformer-based trackers like STARK [6], AiATrack [25], TransT [18] and TrTr [26]. And fully-transformer based trackers like SwinTrack [27], MixFormer [19], Pro-ContEXT [28] and VideoTrack [29].…”

Section: Transformer-based Trackersmentioning

confidence: 99%

Enhancing Single Object Tracking with Temporal Information

Abboud,

Jafar,

Hamdoun

2024

Preprint

View full text Add to dashboard Cite

This study introduces an enhancement approach for single object tracking (SOT) algorithms, aimingto use temporal information with minimize architectural changes while preserving the original modeldimensions with minimal impact on the speed. Most state-of-the-art approaches prioritize effectivefeature extraction from the object or the search window images, while temporal information is oftenneglected. Additionally, many deep models used in SOTs are time-consuming compared to traditionaltrackers. This study focuses on SwinTrack V2 as our case study due to its absence of temporalinformation utilization and its excellent performance on GOT-10k dataset. The method suggest addinga third input to the model, which is an updated object template. The initial template features arefused with the updated template using an attention block, which effectively preserves the originalmodel dimension. Our additional block improves performance when testing on Got-10k dataset by 0.9% for average overlap AO and by 2.1% for success rate SR0.5 compared to the original tracker,with only a slight impact on speed.

show abstract

Section: Transformer-based Trackersmentioning

confidence: 99%

Enhancing Single Object Tracking with Temporal Information

Abboud,

Jafar,

Hamdoun

2024

Preprint

View full text Add to dashboard Cite

show abstract

“…SiamFC++ [6] and SiamCAR [7] algorithms once again introduced the Anchor-Free strategy from object detection into the tracking field, alleviating the problem of hyperparameter sensitivity and improving tracking accuracy. TransT [8] , STARK [9] , TrDiMP [10] , and AiATrack [11] algorithms introduced Transformers for feature enhancement and fusion on the Siamese network, which greatly improved the tracking performance of the algorithms. However, it is clear that the performance improvement of the above algorithms comes at the expense of tracking speed.…”

Section: Siamese Network-based Object Tracking Algorithmmentioning

confidence: 99%

Lightweight object tracking algorithm based on sparse attention

Wang,

Zhang,

Wei

2023

Second International Conference on Electronic Information Technology (EIT 2023)

View full text Add to dashboard Cite

In recent years, with the widespread application of Transformer in computer vision, Transformer-based object tracking algorithms have made some progress in terms of accuracy. However, the large computation, high resource consumption, and slow inference speed of these algorithms severely limit their practical applications. Specifically, these algorithms struggle to meet the real-time tracking demands of resource-limited scenarios such as mobile devices and drones. Therefore, this paper proposes a pure Siamese-based, lightweight object tracking algorithm based on sparse attention. The proposed algorithm significantly improves tracking speed without significantly sacrificing tracking accuracy, making it suitable for practical resource-limited scenarios while still achieving good tracking performance. The proposed algorithm achieves a success rate of 76.3% and a normalized precision of 82.1% on the TrackingNet dataset, at the same time, it achieves a high inference speed of more than 100FPS, which exceeds the mainstream algorithm

show abstract

“…This tracker leverages the prior knowledge of similarity scores obtained in the early stages and proposes an in-network early candidate elimination module, thereby reducing inference time. To address the issue of inhibited performance improvements due to independent correlation calculations in attention mechanisms, AiATrack [22] introduces an Attention in Attention (AiA) module. This module enhances appropriate correlations and suppresses erroneous correlations by seeking consensus among all relevant vectors.…”

Section: Introductionmentioning

confidence: 99%

Single Object Tracking in Satellite Videos Based on Feature Enhancement and Multi-Level Matching Strategy

Yang,

Pan,

Liu

et al. 2023

Remote Sensing

View full text Add to dashboard Cite

Despite significant advancements in remote sensing object tracking (RSOT) in recent years, achieving accurate and continuous tracking of tiny-sized targets remains a challenging task due to similar object interference and other related issues. In this paper, from the perspective of feature enhancement and a better feature matching strategy, we present a tracker SiamTM specifically designed for RSOT, which is mainly based on a new target information enhancement (TIE) module and a multi-level matching strategy. First, we propose a TIE module to address the challenge of tiny object sizes in satellite videos. The proposed TIE module goes along two spatial directions to capture orientation and position-aware information, respectively, while capturing inter-channel information at the global 2D image level. The TIE module enables the network to extract discriminative features of the targets more effectively from satellite images. Furthermore, we introduce a multi-level matching (MM) module that is better suited for satellite video targets. The MM module firstly embeds the target feature map after ROI Align into each position of the search region feature map to obtain a preliminary response map. Subsequently, the preliminary response map and the template region feature map are subjected to the Depth-wise Cross Correlation operation to get a more refined response map. Through this coarse-to-fine approach, the tracker obtains a response map with a more accurate position, which lays a good foundation for the prediction operation of the subsequent sub-networks. We conducted extensive experiments on two large satellite video single-object tracking datasets: SatSOT and SV248S. Without bells and whistles, the proposed tracker SiamTM achieved competitive results on both datasets while running at real-time speed.

show abstract

AiATrack: Attention in Attention for Transformer Visual Tracking

Cited by 3 publications

References 61 publications

Enhancing Single Object Tracking with Temporal Information

Enhancing Single Object Tracking with Temporal Information

Lightweight object tracking algorithm based on sparse attention

Single Object Tracking in Satellite Videos Based on Feature Enhancement and Multi-Level Matching Strategy

Contact Info

Product

Resources

About