arXiv subjects

Jin Tang

Publications and source records attributed to Jin Tang.

At least 163 records · Page 9Linked to original sources

arXiv2023

MutualFormer: Multi-Modality Representation Learning via Cross-Diffusion Attention

Aggregating multi-modality data to obtain reliable data representation attracts more and more attention. Recent studies demonstrate that Transformer models usually work well for multi-modality tasks. Existing Transformers generally either adopt the Cross-Attention (CA) mechanism or simple concatenation to achieve the information interaction among different modalities which generally ignore the issue of modality gap. In this work, we re-think Transformer and extend it to MutualFormer for multi-modality data representation. Rather than CA in Transformer, MutualFormer employs our new design of Cross-Diffusion Attention (CDA) to conduct the information communication among different modalities. Comparing with CA, the main advantages of the proposed CDA are three aspects. First, the crossaffinities in CDA are defined based on the individual modality affinities in the metric space which thus can naturally avoid the issue of modality/domain gap in feature based CA definition. Second, CDA provides a general scheme which can either be used for multimodality representation or serve as the post-optimization for existing CA models. Third, CDA is implemented efficiently. We successfully apply the MutualFormer on different multi-modality learning tasks (i.e., RGB-Depth SOD, RGB-NIR object ReID). Extensive experiments demonstrate the effectiveness of the proposed MutualFormer.

cs.CV↗

arXiv2023

Meta-DMoE: Adapting to Domain Shift by Meta-Distillation from Mixture-of-Experts

In this paper, we tackle the problem of domain shift. Most existing methods perform training on multiple source domains using a single model, and the same trained model is used on all unseen target domains. Such solutions are sub-optimal as each target domain exhibits its own specialty, which is not adapted. Furthermore, expecting single-model training to learn extensive knowledge from multiple source domains is counterintuitive. The model is more biased toward learning only domain-invariant features and may result in negative knowledge transfer. In this work, we propose a novel framework for unsupervised test-time adaptation, which is formulated as a knowledge distillation process to address domain shift. Specifically, we incorporate Mixture-of-Experts (MoE) as teachers, where each expert is separately trained on different source domains to maximize their specialty. Given a test-time target domain, a small set of unlabeled data is sampled to query the knowledge from MoE. As the source domains are correlated to the target domains, a transformer-based aggregator then combines the domain knowledge by examining the interconnection among them. The output is treated as a supervision signal to adapt a student prediction network toward the target domain. We further employ meta-learning to enforce the aggregator to distill positive knowledge and the student network to achieve fast adaptation. Extensive experiments demonstrate that the proposed method outperforms the state-of-the-art and validates the effectiveness of each proposed component. Our code is available at https://github.com/n3il666/Meta-DMoE.

cs.LG↗

arXiv2022

An end-to-end multi-scale network for action prediction in videos

In this paper, we develop an efficient multi-scale network to predict action classes in partial videos in an end-to-end manner. Unlike most existing methods with offline feature generation, our method directly takes frames as input and further models motion evolution on two different temporal scales.Therefore, we solve the complexity problems of the two stages of modeling and the problem of insufficient temporal and spatial information of a single scale. Our proposed End-to-End MultiScale Network (E2EMSNet) is composed of two scales which are named segment scale and observed global scale. The segment scale leverages temporal difference over consecutive frames for finer motion patterns by supplying 2D convolutions. For observed global scale, a Long Short-Term Memory (LSTM) is incorporated to capture motion features of observed frames. Our model provides a simple and efficient modeling framework with a small computational cost. Our E2EMSNet is evaluated on three challenging datasets: BIT, HMDB51, and UCF101. The extensive experiments demonstrate the effectiveness of our method for action prediction in videos.

cs.CV↗

arXiv2022

Combined Magnetic Imaging and Anisotropic Magnetoresistance Detection of Dipolar Skyrmions

Magnetic skyrmions are localized particle-like nontrivial swirls that are promising in building high-performance topological spintronic devices. The read-out functions in skyrmionic devices require the translation of magnetic skyrmions to electrical signals. Here, we report combined real-space magnetic imaging and anisotropic magnetoresistance studies on dipolar skyrmions. A single skyrmion chain and single skyrmion are observed using Lorentz transmission electron microscopy imaging. Meanwhile, the field, helicity, and skyrmion count dependence of anisotropic magnetoresistance of the Fe3Sn2 nanostructures are obtained simultaneously. Our results demonstrate that the anisotropic magnetoresistance of skyrmions is independent of the helicity and proportional to the skyrmion count. Our work could promote read-out operations in skyrmion-based spintronic devices.

cond-mat.mes-hall↗

arXiv2022

Detecting Rotated Objects as Gaussian Distributions and Its 3-D Generalization

Existing detection methods commonly use a parameterized bounding box (BBox) to model and detect (horizontal) objects and an additional rotation angle parameter is used for rotated objects. We argue that such a mechanism has fundamental limitations in building an effective regression loss for rotation detection, especially for high-precision detection with high IoU (e.g. 0.75). Instead, we propose to model the rotated objects as Gaussian distributions. A direct advantage is that our new regression loss regarding the distance between two Gaussians e.g. Kullback-Leibler Divergence (KLD), can well align the actual detection performance metric, which is not well addressed in existing methods. Moreover, the two bottlenecks i.e. boundary discontinuity and square-like problem also disappear. We also propose an efficient Gaussian metric-based label assignment strategy to further boost the performance. Interestingly, by analyzing the BBox parameters' gradients under our Gaussian-based KLD loss, we show that these parameters are dynamically updated with interpretable physical meaning, which help explain the effectiveness of our approach, especially for high-precision detection. We extend our approach from 2-D to 3-D with a tailored algorithm design to handle the heading estimation, and experimental results on twelve public datasets (2-D/3-D, aerial/text/face images) with various base detectors show its superiority.

cs.CV↗

arXiv2022

Beyond Greedy Search: Tracking by Multi-Agent Reinforcement Learning-based Beam Search

To track the target in a video, current visual trackers usually adopt greedy search for target object localization in each frame, that is, the candidate region with the maximum response score will be selected as the tracking result of each frame. However, we found that this may be not an optimal choice, especially when encountering challenging tracking scenarios such as heavy occlusion and fast motion. To address this issue, we propose to maintain multiple tracking trajectories and apply beam search strategy for visual tracking, so that the trajectory with fewer accumulated errors can be identified. Accordingly, this paper introduces a novel multi-agent reinforcement learning based beam search tracking strategy, termed BeamTracking. It is mainly inspired by the image captioning task, which takes an image as input and generates diverse descriptions using beam search algorithm. Accordingly, we formulate the tracking as a sample selection problem fulfilled by multiple parallel decision-making processes, each of which aims at picking out one sample as their tracking result in each frame. Each maintained trajectory is associated with an agent to perform the decision-making and determine what actions should be taken to update related information. When all the frames are processed, we select the trajectory with the maximum accumulated score as the tracking result. Extensive experiments on seven popular tracking benchmark datasets validated the effectiveness of the proposed algorithm.

cs.CV↗

arXiv2022

Error-Aware Spatial Ensembles for Video Frame Interpolation

Video frame interpolation~(VFI) algorithms have improved considerably in recent years due to unprecedented progress in both data-driven algorithms and their implementations. Recent research has introduced advanced motion estimation or novel warping methods as the means to address challenging VFI scenarios. However, none of the published VFI works considers the spatially non-uniform characteristics of the interpolation error (IE). This work introduces such a solution. By closely examining the correlation between optical flow and IE, the paper proposes novel error prediction metrics that partition the middle frame into distinct regions corresponding to different IE levels. Building upon this IE-driven segmentation, and through the use of novel error-controlled loss functions, it introduces an ensemble of spatially adaptive interpolation units that progressively processes and integrates the segmented regions. This spatial ensemble results in an effective and computationally attractive VFI solution. Extensive experimentation on popular video interpolation benchmarks indicates that the proposed solution outperforms the current state-of-the-art (SOTA) in applications of current interest.

cs.CV↗

arXiv2022

Few-Shot Class-Incremental Learning via Entropy-Regularized Data-Free Replay

Few-shot class-incremental learning (FSCIL) has been proposed aiming to enable a deep learning system to incrementally learn new classes with limited data. Recently, a pioneer claims that the commonly used replay-based method in class-incremental learning (CIL) is ineffective and thus not preferred for FSCIL. This has, if truth, a significant influence on the fields of FSCIL. In this paper, we show through empirical results that adopting the data replay is surprisingly favorable. However, storing and replaying old data can lead to a privacy concern. To address this issue, we alternatively propose using data-free replay that can synthesize data by a generator without accessing real data. In observing the the effectiveness of uncertain data for knowledge distillation, we impose entropy regularization in the generator training to encourage more uncertain examples. Moreover, we propose to relabel the generated data with one-hot-like labels. This modification allows the network to learn by solely minimizing the cross-entropy loss, which mitigates the problem of balancing different objectives in the conventional knowledge distillation approach. Finally, we show extensive experimental results and analysis on CIFAR-100, miniImageNet and CUB-200 to demonstrate the effectiveness of our proposed one.

cs.CV↗

arXiv2022

Current-Controlled Topological Magnetic Transformations in a Nanostructured Kagome Magnet

Topological magnetic charge Q is a fundamental parameter that describes the magnetic domains and determines their intriguing electromagnetic properties. The ability to switch Q in a controlled way by electrical methods allows for flexible manipulation of electromagnetic behavior in future spintronic devices. Here we report the room-temperature current-controlled topological magnetic transformations between Q = -1 skyrmions and Q = 0 stripes or type-II bubbles in a kagome crystal Fe$_3$Sn$_2$. We show that the reproducible and reversible skyrmion-bubble and skyrmion-stripe transformations can be achieved by tuning the density of nanosecond pulsed current of the order of ~10$^{10}$ A$^{-2}$. Further numerical simulations suggest that spin-transfer torque combined with Joule thermal heating effects determine the current-induced topological magnetic transformations.

cond-mat.mes-hall↗

arXiv2022

Two-dimensional characterization of three-dimensional magnetic bubbles in Fe$_3$Sn$_2$ nanostructures

We report differential phase contrast scanning transmission electron microscopy (TEM) of nanoscale magnetic objects in Kagome ferromagnet Fe$_3$Sn$_2$ nanostructures. This technique can directly detect the deflection angle of a focused electron beam, thus allowing clear identification of the real magnetic structures of two magnetic objects including three-ring and complex arch-shaped vortices in Fe$_3$Sn$_2$ by Lorentz transmission electron microscopy imaging. Numerical calculations based on real material-specific parameters well reproduced the experimental results, showing that the magnetic objects can be attributed to integral magnetizations of two types of complex three-dimensional (3D) magnetic skyrmion bubbles with depth-modulated spin twisting. Magnetic configurations obtained using the high-resolution TEM are generally considered as two-dimensional (2D) magnetic objects previously. Our results imply the importance of the integral magnetizations of underestimated 3D magnetic structures in 2D TEM magnetic characterizations.

cond-mat.mtrl-sci↗

arXiv2022

Past and Future Motion Guided Network for Audio Visual Event Localization

In recent years, audio-visual event localization has attracted much attention. It's purpose is to detect the segment containing audio-visual events and recognize the event category from untrimmed videos. Existing methods use audio-guided visual attention to lead the model pay attention to the spatial area of the ongoing event, devoting to the correlation between audio and visual information but ignoring the correlation between audio and spatial motion. We propose a past and future motion extraction (pf-ME) module to mine the visual motion from videos ,embedded into the past and future motion guided network (PFAGN), and motion guided audio attention (MGAA) module to achieve focusing on the information related to interesting events in audio modality through the past and future visual motion. We choose AVE as the experimental verification dataset and the experiments show that our method outperforms the state-of-the-arts in both supervised and weakly-supervised settings.

cs.CV↗

arXiv2022

Current-driven Dynamics of Skyrmion Bubbles in Achiral Uniaxial Magnets

We report dynamics of skyrmion bubbles driven by spin-transfer torque in achiral ferromagnetic nanostripes using micromagnetic simulations. In a three-dimensional uniaxial ferromagnet with a quality factor that is smaller than 1, the skyrmion bubble is forced to stay at the central nanostripe by a repulsive force from the geometry border. The coherent motion of skyrmion bubbles in the nanostripe can be realized by increasing the quality factor to ~3.8. Our results should propel the design for future spintronic devices such as artificial neural computing and racetrack memory based on dipole-stabilized skyrmion bubbles.

cond-mat.mtrl-sci↗

arXiv2022

Duality-Gated Mutual Condition Network for RGBT Tracking

Low-quality modalities contain not only a lot of noisy information but also some discriminative features in RGBT tracking. However, the potentials of low-quality modalities are not well explored in existing RGBT tracking algorithms. In this work, we propose a novel duality-gated mutual condition network to fully exploit the discriminative information of all modalities while suppressing the effects of data noise. In specific, we design a mutual condition module, which takes the discriminative information of a modality as the condition to guide feature learning of target appearance in another modality. Such module can effectively enhance target representations of all modalities even in the presence of low-quality modalities. To improve the quality of conditions and further reduce data noise, we propose a duality-gated mechanism and integrate it into the mutual condition module. To deal with the tracking failure caused by sudden camera motion, which often occurs in RGBT tracking, we design a resampling strategy based on optical flow algorithms. It does not increase much computational cost since we perform optical flow calculation only when the model prediction is unreliable and then execute resampling when the sudden camera motion is detected. Extensive experiments on four RGBT tracking benchmark datasets show that our method performs favorably against the state-of-the-art tracking algorithms

cs.CV↗

arXiv2022

SCRDet++: Detecting Small, Cluttered and Rotated Objects via Instance-Level Feature Denoising and Rotation Loss Smoothing

Small and cluttered objects are common in real-world which are challenging for detection. The difficulty is further pronounced when the objects are rotated, as traditional detectors often routinely locate the objects in horizontal bounding box such that the region of interest is contaminated with background or nearby interleaved objects. In this paper, we first innovatively introduce the idea of denoising to object detection. Instance-level denoising on the feature map is performed to enhance the detection to small and cluttered objects. To handle the rotation variation, we also add a novel IoU constant factor to the smooth L1 loss to address the long standing boundary problem, which to our analysis, is mainly caused by the periodicity of angular (PoA) and exchangeability of edges (EoE). By combing these two features, our proposed detector is termed as SCRDet++. Extensive experiments are performed on large aerial images public datasets DOTA, DIOR, UCAS-AOD as well as natural image dataset COCO, scene text dataset ICDAR2015, small traffic light dataset BSTLD and our released S$^2$TLD by this paper. The results show the effectiveness of our approach. The released dataset S2TLD is made public available, which contains 5,786 images with 14,130 traffic light instances across five categories.

cs.CV↗

arXiv2022

Tiny Object Tracking: A Large-scale Dataset and A Baseline

Tiny objects, frequently appearing in practical applications, have weak appearance and features, and receive increasing interests in meany vision tasks, such as object detection and segmentation. To promote the research and development of tiny object tracking, we create a large-scale video dataset, which contains 434 sequences with a total of more than 217K frames. Each frame is carefully annotated with a high-quality bounding box. In data creation, we take 12 challenge attributes into account to cover a broad range of viewpoints and scene complexities, and annotate these attributes for facilitating the attribute-based performance analysis. To provide a strong baseline in tiny object tracking, we propose a novel Multilevel Knowledge Distillation Network (MKDNet), which pursues three-level knowledge distillations in a unified framework to effectively enhance the feature representation, discrimination and localization abilities in tracking tiny objects. Extensive experiments are performed on the proposed dataset, and the results prove the superiority and effectiveness of MKDNet compared with state-of-the-art methods. The dataset, the algorithm code, and the evaluation code are available at https://github.com/mmic-lcl/Datasets-and-benchmark-code.

cs.CV↗

arXiv2022

On the Robustness of "Robust reversible data hiding scheme based on two-layer embedding strategy"

In the paper "Robust reversible data hiding scheme based on two-layer embedding strategy" published in INS recently, Kumar et al. proposed a robust reversible data hiding (RRDH) scheme based on two-layer embedding. Secret data was embedded into the most significant bit (MSB) planes to increase robustness, and a sorting strategy based on local complexity was adopted to reduce distortion. However, Kumar et al.'s reversible data hiding (RDH) scheme is not as robust against joint photographic experts group (JPEG) compression as stated and can not be called RRDH. This comment first gives a brief description of their RDH scheme, then analyses their scheme's robustness from the perspective of JPEG compression principles. JPEG compression will change pixel values, thereby destroying auxiliary information and pixel value ordering required to extract secret data correctly, making their scheme not robust. Next, the changes in both bit plane and pixel value ordering after JPEG compression are shown and analysed by different robustness-testing experiments. Finally, some suggestions are given to improve the robustness.

cs.MM↗

arXiv2021

Magnetic and magnetocaloric effect in the stuffed honeycomb antiferromagnet GdInO3 polycrystalline

The magnetic and magnetocaloric (MCE) properties were studied in a stuffed honeycomb antiferromagnet GdInO3 polycrystalline. No long-range magnetic ordering was observed with only a sharp upturn in the temperature dependent magnetization curves at TN ~ 2.1 K. The large value of frustration index value ~ 5.0 suggests short-range antiferromagnetic interactions existing between the Gd3+ moments in this frustrated magnetic system. Negligible thermal and magnetic hysteresis suggest a second-order feature of phase transition and a reversible magnetocaloric effect (MCE) in GdInO3 compound. In the magnetic field changes of 0-50 kOe and 0-70 kOe, the maximum magnetic entropy change values are 9.31 J/kg K and 17.53 J/kg K near the liquid helium temperature, with the corresponding RCP values of 106.61 and 196.38 J/kg, respectively. The relative lower MCE performance of GdInO3 polycrystalline than the other Gd-based magnetocaloric effect is understood by the high magnetic frustration in this system. Our investigation results reveal GdInO3 polycrystalline has a large reversible MCE, which not only provides another possibility of exploiting magnetocaloric refrigerants in the frustrated magnetic systems near the cryogenic temperature region, but also serves to excavate more exotic properties in the frustrated stuffed honeycomb magnetic systems.

cond-mat.str-el↗

arXiv2021

LasHeR: A Large-scale High-diversity Benchmark for RGBT Tracking

RGBT tracking receives a surge of interest in the computer vision community, but this research field lacks a large-scale and high-diversity benchmark dataset, which is essential for both the training of deep RGBT trackers and the comprehensive evaluation of RGBT tracking methods. To this end, we present a Large-scale High-diversity benchmark for RGBT tracking (LasHeR) in this work. LasHeR consists of 1224 visible and thermal infrared video pairs with more than 730K frame pairs in total. Each frame pair is spatially aligned and manually annotated with a bounding box, making the dataset well and densely annotated. LasHeR is highly diverse capturing from a broad range of object categories, camera viewpoints, scene complexities and environmental factors across seasons, weathers, day and night. We conduct a comprehensive performance evaluation of 12 RGBT tracking algorithms on the LasHeR dataset and present detailed analysis to clarify the research room in RGBT tracking. In addition, we release the unaligned version of LasHeR to attract the research interest for alignment-free RGBT tracking, which is a more practical task in real-world applications. The datasets and evaluation protocols are available at: https://github.com/BUGPLEASEOUT/LasHeR.

cs.CV↗