Breaking Modality Gap in RGBT Tracking: Coupled Knowledge Distillation

• URL: http://arxiv.org/abs/2410.11586v1
• Date: Tue, 15 Oct 2024 13:22:58 GMT
• Title: Breaking Modality Gap in RGBT Tracking: Coupled Knowledge Distillation
• Authors: Andong Lu, Jiacong Zhao, Chenglong Li, Yun Xiao, Bin Luo,
• Abstract summary: Modality gap between RGB and thermal infrared (TIR) images is a crucial issue but often overlooked in existing RGBT tracking methods. We propose a novel Coupled Knowledge Distillation framework called CKD, which pursues common styles of different modalities to break modality gap. In particular, we introduce two student networks and employ the style distillation loss to make their style features consistent.
• Score: 21.161244379091833
• License: http://creativecommons.org/licenses/by-nc-nd/4.0/
• Abstract: Modality gap between RGB and thermal infrared (TIR) images is a crucial issue but often overlooked in existing RGBT tracking methods. It can be observed that modality gap mainly lies in the image style difference. In this work, we propose a novel Coupled Knowledge Distillation framework called CKD, which pursues common styles of different modalities to break modality gap, for high performance RGBT tracking. In particular, we introduce two student networks and employ the style distillation loss to make their style features consistent as much as possible. Through alleviating the style difference of two student networks, we can break modality gap of different modalities well. However, the distillation of style features might harm to the content representations of two modalities in student networks. To handle this issue, we take original RGB and TIR networks as the teachers, and distill their content knowledge into two student networks respectively by the style-content orthogonal feature decoupling scheme. We couple the above two distillation processes in an online optimization framework to form new feature representations of RGB and thermal modalities without modality gap. In addition, we design a masked modeling strategy and a multi-modal candidate token elimination strategy into CKD to improve tracking robustness and efficiency respectively. Extensive experiments on five standard RGBT tracking datasets validate the effectiveness of the proposed method against state-of-the-art methods while achieving the fastest tracking speed of 96.4 FPS. Code available at https://github.com/Multi-Modality-Tracking/CKD.

Related papers

• UHKD: A Unified Framework for Heterogeneous Knowledge Distillation via Frequency-Domain Representations [5.382357091398666]
  Unified Heterogeneous Knowledge Distillation (UHKD) is proposed as a framework that leverages intermediate features in the frequency domain for cross-architecture transfer.<n>Experiments on CIFAR-100 and ImageNet-1K demonstrate gains of 5.59% and 0.83% over the latest method.
  arXiv  Detail & Related papers  (2025-10-28T06:41:43Z)
• Lightweight Facial Landmark Detection in Thermal Images via Multi-Level Cross-Modal Knowledge Transfer [13.887803692033073]
  Facial Landmark Detection in thermal imagery is critical for applications in challenging lighting conditions.<n>We propose a novel framework that decouples high-fidelity RGB-to-thermal knowledge transfer from model compression.<n> Experiments show that our approach sets a new state-of-the-art on public thermal FLD benchmarks, notably outperforming previous methods.
  arXiv  Detail & Related papers  (2025-10-13T08:19:56Z)
• Multi-Modal Hybrid Learning and Sequential Training for RGB-T Saliency Detection [10.589062261564631]
  RGB-T saliency detection has emerged as an important computer vision task, identifying conspicuous objects in challenging scenes such as dark environments. Existing methods neglect the characteristics of cross-modal features and rely solely on network structures to fuse RGB and thermal features. We first propose a Multi-Modal Hybrid loss (MMHL) that comprises supervised and self-supervised loss functions.
  arXiv  Detail & Related papers  (2023-09-13T20:47:29Z)
• Learning Multi-Granular Spatio-Temporal Graph Network for Skeleton-based Action Recognition [49.163326827954656]
  We propose a novel multi-granular-temporal graph network for skeleton-based action classification. We develop a dual-head graph network consisting of two inter-leaved branches, which enables us to extract at least two-temporal resolutions. We conduct extensive experiments on three large-scale datasets.
  arXiv  Detail & Related papers  (2021-08-10T09:25:07Z)
• SFANet: A Spectrum-aware Feature Augmentation Network for Visible-Infrared Person Re-Identification [12.566284647658053]
  We propose a novel spectrum-aware feature augementation network named SFANet for cross-modality matching problem. Learning with grayscale-spectrum images, our model can apparently reduce modality discrepancy and detect inner structure relations. In feature-level, we improve the conventional two-stream network through balancing the number of specific and sharable convolutional blocks.
  arXiv  Detail & Related papers  (2021-02-24T08:57:32Z)
• RGBT Tracking via Multi-Adapter Network with Hierarchical Divergence Loss [37.99375824040946]
  We propose a novel multi-adapter network to jointly perform modality-shared, modality-specific and instance-aware target representation learning. Experiments on two RGBT tracking benchmark datasets demonstrate the outstanding performance of the proposed tracker.
  arXiv  Detail & Related papers  (2020-11-14T01:50:46Z)
• Searching Multi-Rate and Multi-Modal Temporal Enhanced Networks for Gesture Recognition [89.0152015268929]
  We propose the first neural architecture search (NAS)-based method for RGB-D gesture recognition. The proposed method includes two key components: 1) enhanced temporal representation via the 3D Central Difference Convolution (3D-CDC) family, and optimized backbones for multi-modal-rate branches and lateral connections. The resultant multi-rate network provides a new perspective to understand the relationship between RGB and depth modalities and their temporal dynamics.
  arXiv  Detail & Related papers  (2020-08-21T10:45:09Z)
• Dynamic Dual-Attentive Aggregation Learning for Visible-Infrared Person Re-Identification [208.1227090864602]
  Visible-infrared person re-identification (VI-ReID) is a challenging cross-modality pedestrian retrieval problem. Existing VI-ReID methods tend to learn global representations, which have limited discriminability and weak robustness to noisy images. We propose a novel dynamic dual-attentive aggregation (DDAG) learning method by mining both intra-modality part-level and cross-modality graph-level contextual cues for VI-ReID.
  arXiv  Detail & Related papers  (2020-07-18T03:08:13Z)
• Bi-directional Cross-Modality Feature Propagation with Separation-and-Aggregation Gate for RGB-D Semantic Segmentation [59.94819184452694]
  Depth information has proven to be a useful cue in the semantic segmentation of RGBD images for providing a geometric counterpart to the RGB representation. Most existing works simply assume that depth measurements are accurate and well-aligned with the RGB pixels and models the problem as a cross-modal feature fusion. In this paper, we propose a unified and efficient Crossmodality Guided to not only effectively recalibrate RGB feature responses, but also to distill accurate depth information via multiple stages and aggregate the two recalibrated representations alternatively.
  arXiv  Detail & Related papers  (2020-07-17T18:35:24Z)
• Temporal Distinct Representation Learning for Action Recognition [139.93983070642412]
  Two-Dimensional Convolutional Neural Network (2D CNN) is used to characterize videos. Different frames of a video share the same 2D CNN kernels, which may result in repeated and redundant information utilization. We propose a sequential channel filtering mechanism to excite the discriminative channels of features from different frames step by step, and thus avoid repeated information extraction. Our method is evaluated on benchmark temporal reasoning datasets Something-Something V1 and V2, and it achieves visible improvements over the best competitor by 2.4% and 1.3%, respectively.
  arXiv  Detail & Related papers  (2020-07-15T11:30:40Z)
• Multi-interactive Dual-decoder for RGB-thermal Salient Object Detection [37.79290349045164]
  RGB-thermal salient object detection (SOD) aims to segment the common prominent regions of visible image and corresponding thermal infrared image. Existing methods don't fully explore and exploit the potentials of complementarity of different modalities and multi-type cues of image contents. We propose a multi-interactive dual-decoder to mine and model the multi-type interactions for accurate RGBT SOD.
  arXiv  Detail & Related papers  (2020-05-05T16:21:17Z)
• Unpaired Multi-modal Segmentation via Knowledge Distillation [77.39798870702174]
  We propose a novel learning scheme for unpaired cross-modality image segmentation. In our method, we heavily reuse network parameters, by sharing all convolutional kernels across CT and MRI. We have extensively validated our approach on two multi-class segmentation problems.
  arXiv  Detail & Related papers  (2020-01-06T20:03:17Z)

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.