Dual-branch Distilled Transformer for Efficient Asymmetric UAV Tracking

• URL: http://arxiv.org/abs/2605.28018v1
• Date: Wed, 27 May 2026 06:20:13 GMT
• Title: Dual-branch Distilled Transformer for Efficient Asymmetric UAV Tracking
• Abstract summary: EATrack is a teacher-guided dual-branch distillation strategy for UAV tracking.<n>It compensates for weakened representations by guiding the student to learn strong target representations.<n>Experiments on five UAV benchmarks demonstrate that EATrack achieves a favorable balance between accuracy and speed.
• Score: 34.13651852483772
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Given the real-time demands of UAV tracking, many methods simplify the backbone to reduce computation, but this often weakens feature representation and degrades performance in complex scenarios. To alleviate this issue, we propose EATrack, an efficient and asymmetric UAV tracking framework centered around a teacher-guided dual-branch distillation strategy that enhances the feature expressiveness of the lightweight student model. Specifically, EATrack investigates two complementary perspectives of knowledge transfer: spatially focused feature-level distillation that compensates for weakened representations by guiding the student to learn strong target representations, and prediction-level distillation that enhances spatial localization by learning the teacher's capability for accurate target localization. Furthermore, to enhance robustness against appearance variations, we introduce a fine-grained target-aware distillation strategy that selectively transfers the teacher's target modeling capacity to the student. A temporal adaptation module is incorporated at inference to enhance robustness over time. Experiments on five UAV benchmarks demonstrate that EATrack achieves a favorable balance between accuracy and speed. Code: https://github.com/GXNU-ZhongLab/EATrack

Related papers

• Towards Compact Unified Multimodal Tracking: Synergizing Knowledge Distillation with Structural Pruning [80.73561460934809]
  multimodal object tracking has achieved remarkable robustness by leveraging complementary sensor data.<n>We identify the prediction head as a critical but often overlooked efficiency bottleneck.<n>By strategically streamlining the decoder architecture, we unlock the potential for real-time inference.
  arXiv  Detail & Related papers  (2026-08-02T20:41:33Z)
• VISD: Enhancing Video Reasoning via Structured Self-Distillation [4.8876574457819855]
  Training VideoLLMs for complex reasoning remains challenging due to sequence of level level rewards and the lack of fine grained credit assignment over long, temporally grounded reasoning trajectories.<n>We propose VISD, a structured self distillation framework that introduces diagnostically meaningful privileged information for video reasoning.
  arXiv  Detail & Related papers  (2026-05-07T12:13:15Z)
• EvoDriveVLA: Evolving Autonomous Driving Vision-Language-Action Model via Collaborative Perception-Planning Distillation [58.84721000276226]
  EvoDriveVLA is a novel collaborative perception-planning framework.<n>It integrates self-anchored perceptual constraints and oracle-guided trajectory optimization.<n>EvoDriveVLA achieves SOTA performance in open-loop evaluation and significantly enhances performance in closed-loop evaluation.
  arXiv  Detail & Related papers  (2026-03-10T10:19:07Z)
• TopoCurate:Modeling Interaction Topology for Tool-Use Agent Training [53.93696896939915]
  Training tool-use agents typically rely on Supervised Fine-Tuning (SFT) on successful trajectories and Reinforcement Learning (RL) on pass-rate-selected tasks.<n>We propose TopoCurate, an interaction-aware framework that projects multi-trial rollouts from the same task into a unified semantic quotient topology.<n>TopoCurate achieves consistent gains of 4.2% (SFT) and 6.9% (RL) over state-of-the-art baselines.
  arXiv  Detail & Related papers  (2026-03-02T10:38:54Z)
• Enhancing Training Data Attribution with Representational Optimization [57.61977909113113]
  Training data attribution methods aim to measure how training data impacts a model's predictions.<n>We propose AirRep, a representation-based approach that closes this gap by learning task-specific and model-aligned representations explicitly for TDA.<n>AirRep introduces two key innovations: a trainable encoder tuned for attribution quality, and an attention-based pooling mechanism that enables accurate estimation of group-wise influence.
  arXiv  Detail & Related papers  (2025-05-24T05:17:53Z)
• Learning Occlusion-Robust Vision Transformers for Real-Time UAV Tracking [11.146155422858824]
  Single-stream architectures using Vision Transformer (ViT) backbones show great potential for real-time UAV tracking.<n>We propose to learn Occlusion-Robust Representations (ORR) based on ViTs for UAV tracking.<n>We also propose an Adaptive Feature-Based Knowledge Distillation (AFKD) method to create a more compact tracker.
  arXiv  Detail & Related papers  (2025-04-12T14:06:50Z)
• Learning an Adaptive and View-Invariant Vision Transformer for Real-Time UAV Tracking [15.597151507814429]
  Transformer-based models have improved visual tracking, but most still cannot run in real time on resource devices.<n>We propose AVTrack, an adaptive tracking framework that adaptively activates transformer blocks through an Activation Module (AM)<n>To address extreme viewpoint variations, we propose to learn view-in representations via mutual information (MI)<n>In addition, we propose AVTrack-MD, an enhanced tracker incorporating a novel MI-based multi-teacher knowledge framework.
  arXiv  Detail & Related papers  (2024-12-28T03:57:44Z)
• Efficient Diffusion as Low Light Enhancer [63.789138528062225]
  Reflectance-Aware Trajectory Refinement (RATR) is a simple yet effective module to refine the teacher trajectory using the reflectance component of images. textbfReflectance-aware textbfDiffusion with textbfDistilled textbfTrajectory (textbfReDDiT) is an efficient and flexible distillation framework tailored for Low-Light Image Enhancement (LLIE)
  arXiv  Detail & Related papers  (2024-10-16T08:07:18Z)
• General Compression Framework for Efficient Transformer Object Tracking [29.780411151321438]
  We propose a general model compression framework for efficient transformer object tracking, named CompressTracker.<n>Our approach features a novel stage division strategy that segments the transformer layers of the teacher model into distinct stages.<n>We also design a unique replacement training technique that randomly substitutes specific stages in the student model with those from the teacher model.
  arXiv  Detail & Related papers  (2024-09-26T06:27:15Z)
• Progressive Representation Learning for Real-Time UAV Tracking [20.76053366492599]
  This work proposes a novel progressive representation learning framework for UAV tracking, i.e., PRL-Track. For coarse representation learning, two innovative regulators, which rely on appearance and semantic information, are designed to mitigate appearance interference and capture semantic information. For fine representation learning, a new hierarchical modeling generator is developed to intertwine coarse object representations.
  arXiv  Detail & Related papers  (2024-09-25T06:16:32Z)
• Promoting CNNs with Cross-Architecture Knowledge Distillation for Efficient Monocular Depth Estimation [4.242540533823568]
  Transformer models are usually computationally-expensive, and their effectiveness in light-weight models are limited compared to convolutions. We propose a cross-architecture knowledge distillation method for MDE, dubbed DisDepth, to enhance efficient CNN models with the supervision of state-of-the-art transformer models. Our method achieves significant improvements on various efficient backbones, showcasing its potential for efficient monocular depth estimation.
  arXiv  Detail & Related papers  (2024-04-25T07:55:47Z)
• Towards Discriminative Representations with Contrastive Instances for Real-Time UAV Tracking [5.557099240958562]
  Discriminative correlation filters (DCF)-based trackers can yield high efficiency on a single CPU but with inferior precision. Lightweight Deep learning (DL)-based trackers can achieve a good balance between efficiency and precision but performance gains are limited by the compression rate. This paper aims to enhance the discriminative power of feature representations from a new feature-learning perspective.
  arXiv  Detail & Related papers  (2023-08-22T13:58:45Z)
• Self-Promoted Supervision for Few-Shot Transformer [178.52948452353834]
  Self-promoted sUpervisioN (SUN) is a few-shot learning framework for vision transformers (ViTs) SUN pretrains the ViT on the few-shot learning dataset and then uses it to generate individual location-specific supervision for guiding each patch token. Experiments show that SUN using ViTs significantly surpasses other few-shot learning frameworks with ViTs and is the first one that achieves higher performance than those CNN state-of-the-arts.
  arXiv  Detail & Related papers  (2022-03-14T12:53:27Z)
• Visual Alignment Constraint for Continuous Sign Language Recognition [74.26707067455837]
  Vision-based Continuous Sign Language Recognition aims to recognize unsegmented gestures from image sequences. In this work, we revisit the overfitting problem in recent CTC-based CSLR works and attribute it to the insufficient training of the feature extractor. We propose a Visual Alignment Constraint (VAC) to enhance the feature extractor with more alignment supervision.
  arXiv  Detail & Related papers  (2021-04-06T07:24:58Z)

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.