Deep object tracking can be modeled by online learning to adapt to the appearance changes, or offline learning to achieve fast tracking speed. However, either online or offline learning trackers are still difficult to...
详细信息
Deep object tracking can be modeled by online learning to adapt to the appearance changes, or offline learning to achieve fast tracking speed. However, either online or offline learning trackers are still difficult to continuously cope with challenging tracking scenes, due to the cumulative errors easily caused by online learning trackers and the object appearance changes often neglected by offline learning trackers. To overcome this problem, we propose a novel object tracking method based on collaborative framework with adaptive multi-strategy, enabling automatic switch between the online and offline learning trackers. In this framework, we first design the variable action set (VAS) module to adaptively choose the different trackers, tracking strategies, and template strategies with multiple action labels. Moreover, we incorporate the template update and selection (TUS) module, which dynamically updates and selects templates from a memory unit to adapt to object appearance changes. To update the action set and the memory unit, we further devise the online reliability evaluation (ORE) module that not only evaluates the tracking result and the tracker itself, but also estimates the quality and quantity of templates. Comprehensive experiments on challenging short-term and long-term tracking benchmarks demonstrate the remarkable performance of the proposed method.
Recently, CNN and Transformer hybrid networks demonstrated excellent performance in face super-resolution (FSR) tasks. Since numerous features at different scales in hybrid networks, how to fuse these multi-scale feat...
详细信息
Face restoration (FR) is a specialized field within image restoration that aims to recover low-quality (LQ) face images into high-quality (HQ) face images. Recent advances in deep learning technology have led to signi...
详细信息
Point-wise scanning modalities like Optical Coherence Tomography (OCT) or Scanning Laser Ophthalmoscopy suffer from distortions due to the perpetual motion of the eye. While various motion correction approaches have b...
详细信息
The effectiveness of Graph Convolutional Networks (GCNs) has been widely demonstrated in skeleton-based action recognition. However, most existing GCN-based methods use a dense adjacency matrix to describe the structu...
详细信息
Parkinson’s disease (PD) is the 2 nd most prevalent neurodegenerative disease in the world. Thus, the early detection of PD has recently been the subject of several scientific and commercial studies. In this paper, ...
Parkinson’s disease (PD) is the 2 nd most prevalent neurodegenerative disease in the world. Thus, the early detection of PD has recently been the subject of several scientific and commercial studies. In this paper, we propose a pipeline using Vision Transformer applied to mel-spectrograms for PD classification using multilingual sustained vowel recordings. Furthermore, our proposed transformed-based model shows a great potential to use voice as a single modality biomarker for automatic PD detection without language restrictions, a wide range of vowels, with an F1-score equal to 0.78. The results of our study fall within the range of the estimated prevalence of voice and speech disorders in Parkinson’s disease, which ranges from 70-90%. Our study demonstrates a high potential for adaptation in clinical decision-making, allowing for increasingly systematic and fast diagnosis of PD with the potential for use in *** relevance— There is an urgent need to develop non invasive biomarker of Parkinson’s disease effective enough to detect the onset of the disease to introduce neuroprotective treatment at the earliest stage possible and to follow the results of that intervention. Voice disorders in PD are very frequent and are expected to be utilized as an early diagnostic biomarker. The voice analysis using deep neural networks open new opportunities to assess neurodegenerative diseases’ symptoms, for fast diagnosis-making, to guide treatment initiation, and risk prediction. The detection accuracy for voice biomarkers according to our method reached close to the maximum achievable value.
This work reviews the results of the NTIRE 2023 Challenge on Image Shadow Removal. The described set of solutions were proposed for a novel dataset, which captures a wide range of object-light interactions. It consist...
详细信息
Understanding the relationship between vocal tract motion during speech and the resulting acoustic signal is crucial for aided clinical assessment and developing personalized treatment and rehabilitation strategies. T...
详细信息
We investigate the second-order topological phases in a two-dimensional ring resonator array with each plaquette occupied by π gauge flux and imaginary gauge field. The real and imaginary gauge fields are induced by ...
详细信息
We investigate the second-order topological phases in a two-dimensional ring resonator array with each plaquette occupied by π gauge flux and imaginary gauge field. The real and imaginary gauge fields are induced by shifting the displacement and integrating gain or loss into the two half perimeters of the auxiliary rings. The system supports topological corner modes with their emergence being determined by the non-Bloch topological invariant due to skin effects. The bulk modes, exhibiting second-order skin effects in both trivial and nontrivial phases, are accumulated at opposite corners depending on whether clockwise or counterclockwise modes are excited. By introducing an interface with different imaginary gauge fields, we show the bulk modes exist at the interface while the topological corner modes are localized at the physical corners. Furthermore, the skin effects are also presented in the passive ring resonators. The study may find applications in lasers and broadband light trapping.
暂无评论