Dynamic Loss for Robust Learning.

Label noise and class imbalance are common challenges encountered in real-world datasets. Existing approaches for robust learning often focus on addressing either label noise or class imbalance individually, resulting in suboptimal performance when both biases are present. To bridge this gap, this work introduces a novel meta-learning-based dynamic loss that adapts the objective functions during the training process to effectively learn a classifier from long-tailed noisy data. Specifically, our dynamic loss consists of two components: a label corrector and a margin generator. The label corrector is responsible for correcting noisy labels, while the margin generator generates per-class classification margins by capturing the underlying data distribution and the learning state of the classifier. In addition, we employ a hierarchical sampling strategy that enriches a small amount of unbiased metadata with diverse and challenging samples. This enables the joint optimization of the two components in the dynamic loss through meta-learning, allowing the classifier to effectively adapt to clean and balanced test data. Extensive experiments conducted on multiple real-world and synthetic datasets with various types of data biases, including CIFAR-10/100, Animal-10N, ImageNet-LT, and Webvision, demonstrate that our method achieves state-of-the-art accuracy.

MetaSeg: Content-Aware Meta-Net for Omni-Supervised Semantic Segmentation.

Noisy labels, inevitably existing in pseudo-segmentation labels generated from weak object-level annotations, severely hamper model optimization for semantic segmentation. Previous works often rely on massive handcrafted losses and carefully tuned hyperparameters to resist noise, suffering poor generalization capability and high model complexity. Inspired by recent advances in meta-learning, we argue that rather than struggling to tolerate noise hidden behind clean labels passively, a more feasible solution would be to find out the noisy regions actively, so as to simply ignore them during model optimization. With this in mind, this work presents a novel meta-learning-based semantic segmentation method, MetaSeg, that comprises a primary content-aware meta-net (CAM-Net) to serve as a noise indicator for an arbitrary segmentation model counterpart. Specifically, CAM-Net learns to generate pixel-wise weights to suppress noisy regions with incorrect pseudo-labels while highlighting clean ones by exploiting hybrid strengthened features from image content, providing straightforward and reliable guidance for optimizing the segmentation model. Moreover, to break the barrier of time-consuming training when applying meta-learning to common large segmentation models, we further present a new decoupled training strategy that optimizes different model layers in a divide-and-conquer manner. Extensive experiments on object, medical, remote sensing, and human segmentation show that our method achieves superior performance, approaching that of fully supervised settings, which paves a new promising way for omni-supervised semantic segmentation.

SiamATL: Online Update of Siamese Tracking Network via Attentional Transfer Learning.

Visual object tracking with semantic deep features has recently attracted much attention in computer vision. Especially, Siamese trackers, which aim to learn a decision making-based similarity evaluation, are widely utilized in the tracking community. However, the online updating of the Siamese fashion is still a tricky issue due to the limitation, which is a tradeoff between model adaption and degradation. To address such an issue, in this article, we propose a novel attentional transfer learning-based Siamese network (SiamATL), which fully exploits the previous knowledge to inspire the current tracker learning in the decision-making module. First, we explicitly model the template and surroundings by using an attentional online update strategy to avoid template pollution. Then, we introduce an instance-transfer discriminative correlation filter (ITDCF) to enhance the distinguishing ability of the tracker. Finally, we suggest a mutual compensation mechanism that integrates cross-correlation matching and ITDCF detection into the decision-making subnetwork to achieve online tracking. Comprehensive experiments demonstrate that our approach outperforms state-of-the-art tracking algorithms on multiple large-scale tracking datasets.

Physics-based learning with channel attention for Fourier ptychographic microscopy.

Fourier ptychographic microscopy (FPM) is a computational imaging technology for large field-of-view, high resolution and quantitative phase imaging. In FPM, low-resolution intensity images captured with angle-varying illumination are synthesized in Fourier space with phase retrieval approaches. However, system errors such as pupil aberration and light-emitting diode (LED) intensity error seriously affect the reconstruction performance. In this article, we propose a physics-based neural network with channel attention for FPM reconstruction. With the channel attention module, which is introduced into physics-based neural networks for the first time, the spatial distribution of LED intensity can be adaptively corrected. Besides, the channel attention module is used to synthesize different Zernike modes and recover the pupil function. Detailed simulations and experiments are carried out to validate the effectiveness and robustness of the proposed method. The results demonstrate that our method achieves better performance in high-resolution complex field reconstruction, LED intensity correction and pupil function recovery compared with the state-of-art methods. The combination with deep neural network structures like channel attention modules significantly enhance the performance of physics-based neural networks and will promote the application of FPM in practical use.

Moving Object Detection Using Scanning Camera on a High-Precision Intelligent Holder.

During the process of moving object detection in an intelligent visual surveillance system, a scenario with complex background is sure to appear. The traditional methods, such as "frame difference" and "optical flow", may not able to deal with the problem very well. In such scenarios, we use a modified algorithm to do the background modeling work. In this paper, we use edge detection to get an edge difference image just to enhance the ability of resistance illumination variation. Then we use a "multi-block temporal-analyzing LBP (Local Binary Pattern)" algorithm to do the segmentation. In the end, a connected component is used to locate the object. We also produce a hardware platform, the core of which consists of the DSP (Digital Signal Processor) and FPGA (Field Programmable Gate Array) platforms and the high-precision intelligent holder.

Real-Time Tracking Framework with Adaptive Features and Constrained Labels.

This paper proposes a novel tracking framework with adaptive features and constrained labels (AFCL) to handle illumination variation, occlusion and appearance changes caused by the variation of positions. The novel ensemble classifier, including the Forward-Backward error and the location constraint is applied, to get the precise coordinates of the promising bounding boxes. The Forward-Backward error can enhance the adaptation and accuracy of the binary features, whereas the location constraint can overcome the label noise to a certain degree. We use the combiner which can evaluate the online templates and the outputs of the classifier to accommodate the complex situation. Evaluation of the widely used tracking benchmark shows that the proposed framework can significantly improve the tracking accuracy, and thus reduce the processing time. The proposed framework has been tested and implemented on the embedded system using TMS320C6416 and Cyclone Ⅲ kernel processors. The outputs show that achievable and satisfying results can be obtained.