Person re-identification (re-ID) continues to pose a significant challenge, particularly in scenarios involving occlusions. Prior approaches aimed at tackling occlusions have predominantly focused on aligning physical body features through the utilization of external semantic cues. However, these methods tend to be intricate and susceptible to noise. To address the aforementioned challenges, we present an innovative end-to-end solution known as the Dynamic Patch-aware Enrichment Transformer (DPEFormer). This model effectively distinguishes human body information from occlusions automatically and dynamically, eliminating the need for external detectors or precise image alignment. Specifically, we introduce a dynamic patch token selection module (DPSM). DPSM utilizes a label-guided proxy token as an intermediary to identify informative occlusion-free tokens. These tokens are then selected for deriving subsequent local part features. To facilitate the seamless integration of global classification features with the finely detailed local features selected by DPSM, we introduce a novel feature blending module (FBM). FBM enhances feature representation through the complementary nature of information and the exploitation of part diversity. Furthermore, to ensure that DPSM and the entire DPEFormer can effectively learn with only identity labels, we also propose a Realistic Occlusion Augmentation (ROA) strategy. This strategy leverages the recent advances in the Segment Anything Model (SAM). As a result, it generates occlusion images that closely resemble real-world occlusions, greatly enhancing the subsequent contrastive learning process. Experiments on occluded and holistic re-ID benchmarks signify a substantial advancement of DPEFormer over existing state-of-the-art approaches. The code will be made publicly available.
Among the widely used parameter-efficient finetuning (PEFT) methods, LoRA and its variants have gained considerable popularity because of avoiding additional inference costs. However, there still often exists an accuracy gap between these methods and full fine-tuning (FT). In this work, we first introduce a novel weight decomposition analysis to investigate the inherent differences between FT and LoRA. Aiming to resemble the learning capacity of FT from the findings, we propose Weight-Decomposed LowRank Adaptation (DoRA). DoRA decomposes the pre-trained weight into two components, magnitude and direction, for fine-tuning, specifically employing LoRA for directional updates to efficiently minimize the number of trainable parameters. By employing DoRA, we enhance both the learning capacity and training stability of LoRA while avoiding any additional inference overhead. DoRA consistently outperforms LoRA on fine-tuning LLaMA, LLaVA, and VL-BART on various downstream tasks, such as commonsense reasoning, visual instruction tuning, and image/video-text understanding.
A novel online clustering algorithm is presented where an Evolving Restricted Boltzmann Machine (ERBM) is embedded with a Kohonen Network called ERBM-KNet. The proposed ERBM-KNet efficiently handles streaming data in a single-pass mode using the ERBM, employing a bias-variance strategy for neuron growing and pruning, as well as online clustering based on a cluster update strategy for cluster prediction and cluster center update using KNet. Initially, ERBM evolves its architecture while processing unlabeled image data, effectively disentangling the data distribution in the latent space. Subsequently, the KNet utilizes the feature extracted from ERBM to predict the number of clusters and updates the cluster centers. By overcoming the common challenges associated with clustering algorithms, such as prior initialization of the number of clusters and subpar clustering accuracy, the proposed ERBM-KNet offers significant improvements. Extensive experimental evaluations on four benchmarks and one industry dataset demonstrate the superiority of ERBM-KNet compared to state-of-the-art approaches.
Recent advancements in neural compression have surpassed traditional codecs in PSNR and MS-SSIM measurements. However, at low bit-rates, these methods can introduce visually displeasing artifacts, such as blurring, color shifting, and texture loss, thereby compromising perceptual quality of images. To address these issues, this study presents an enhanced neural compression method designed for optimal visual fidelity. We have trained our model with a sophisticated semantic ensemble loss, integrating Charbonnier loss, perceptual loss, style loss, and a non-binary adversarial loss, to enhance the perceptual quality of image reconstructions. Additionally, we have implemented a latent refinement process to generate content-aware latent codes. These codes adhere to bit-rate constraints, balance the trade-off between distortion and fidelity, and prioritize bit allocation to regions of greater importance. Our empirical findings demonstrate that this approach significantly improves the statistical fidelity of neural image compression. On CLIC2024 validation set, our approach achieves a 62% bitrate saving compared to MS-ILLM under FID metric.
The growth of generative adversarial network (GAN) models has increased the ability of image processing and provides numerous industries with the technology to produce realistic image transformations. However, with the field being recently established there are new evaluation metrics that can further this research. Previous research has shown the Fr\'echet Inception Distance (FID) to be an effective metric when testing these image-to-image GANs in real-world applications. Signed Inception Distance (SID), a founded metric in 2023, expands on FID by allowing unsigned distances. This paper uses public datasets that consist of fa\c{c}ades, cityscapes, and maps within Pix2Pix and CycleGAN models. After training these models are evaluated on both inception distance metrics which measure the generating performance of the trained models. Our findings indicate that usage of the metric SID incorporates an efficient and effective metric to complement, or even exceed the ability shown using the FID for the image-to-image GANs
Quantifying cardiovascular parameters like ejection fraction in zebrafish as a host of biological investigations has been extensively studied. Since current manual monitoring techniques are time-consuming and fallible, several image processing frameworks have been proposed to automate the process. Most of these works rely on supervised deep-learning architectures. However, supervised methods tend to be overfitted on their training dataset. This means that applying the same framework to new data with different imaging setups and mutant types can severely decrease performance. We have developed a Zebrafish Automatic Cardiovascular Assessment Framework (ZACAF) to quantify the cardiac function in zebrafish. In this work, we further applied data augmentation, Transfer Learning (TL), and Test Time Augmentation (TTA) to ZACAF to improve the performance for the quantification of cardiovascular function quantification in zebrafish. This strategy can be integrated with the available frameworks to aid other researchers. We demonstrate that using TL, even with a constrained dataset, the model can be refined to accommodate a novel microscope setup, encompassing diverse mutant types and accommodating various video recording protocols. Additionally, as users engage in successive rounds of TL, the model is anticipated to undergo substantial enhancements in both generalizability and accuracy. Finally, we applied this approach to assess the cardiovascular function in nrap mutant zebrafish, a model of cardiomyopathy.
This article explores how emerging generative artificial intelligence (GenAI) models, such as large language models (LLMs), can enhance solution methodologies within process systems engineering (PSE). These cutting-edge GenAI models, particularly foundation models (FMs), which are pre-trained on extensive, general-purpose datasets, offer versatile adaptability for a broad range of tasks, including responding to queries, image generation, and complex decision-making. Given the close relationship between advancements in PSE and developments in computing and systems technologies, exploring the synergy between GenAI and PSE is essential. We begin our discussion with a compact overview of both classic and emerging GenAI models, including FMs, and then dive into their applications within key PSE domains: synthesis and design, optimization and integration, and process monitoring and control. In each domain, we explore how GenAI models could potentially advance PSE methodologies, providing insights and prospects for each area. Furthermore, the article identifies and discusses potential challenges in fully leveraging GenAI within PSE, including multiscale modeling, data requirements, evaluation metrics and benchmarks, and trust and safety, thereby deepening the discourse on effective GenAI integration into systems analysis, design, optimization, operations, monitoring, and control. This paper provides a guide for future research focused on the applications of emerging GenAI in PSE.
Drawing inspiration from the primate brain's intriguing evidence accumulation process, and guided by models from cognitive psychology and neuroscience, the paper introduces the NYCTALE framework, a neuro-inspired and evidence accumulation-based Transformer architecture. The proposed neuro-inspired NYCTALE offers a novel pathway in the domain of Personalized Medicine (PM) for lung cancer diagnosis. In nature, Nyctales are small owls known for their nocturnal behavior, hunting primarily during the darkness of night. The NYCTALE operates in a similarly vigilant manner, i.e., processing data in an evidence-based fashion and making predictions dynamically/adaptively. Distinct from conventional Computed Tomography (CT)-based Deep Learning (DL) models, the NYCTALE performs predictions only when sufficient amount of evidence is accumulated. In other words, instead of processing all or a pre-defined subset of CT slices, for each person, slices are provided one at a time. The NYCTALE framework then computes an evidence vector associated with contribution of each new CT image. A decision is made once the total accumulated evidence surpasses a specific threshold. Preliminary experimental analyses conducted using a challenging in-house dataset comprising 114 subjects. The results are noteworthy, suggesting that NYCTALE outperforms the benchmark accuracy even with approximately 60% less training data on this demanding and small dataset.
Grounded Multimodal Named Entity Recognition (GMNER) is a nascent multimodal task that aims to identify named entities, entity types and their corresponding visual regions. GMNER task exhibits two challenging properties: 1) The weak correlation between image-text pairs in social media results in a significant portion of named entities being ungroundable. 2) There exists a distinction between coarse-grained referring expressions commonly used in similar tasks (e.g., phrase localization, referring expression comprehension) and fine-grained named entities. In this paper, we propose RiVEG, a unified framework that reformulates GMNER into a joint MNER-VE-VG task by leveraging large language models (LLMs) as a connecting bridge. This reformulation brings two benefits: 1) It maintains the optimal MNER performance and eliminates the need for employing object detection methods to pre-extract regional features, thereby naturally addressing two major limitations of existing GMNER methods. 2) The introduction of entity expansion expression and Visual Entailment (VE) Module unifies Visual Grounding (VG) and Entity Grounding (EG). It enables RiVEG to effortlessly inherit the Visual Entailment and Visual Grounding capabilities of any current or prospective multimodal pretraining models. Extensive experiments demonstrate that RiVEG outperforms state-of-the-art methods on the existing GMNER dataset and achieves absolute leads of 10.65%, 6.21%, and 8.83% in all three subtasks.
Zero-shot incremental learning aims to enable the model to generalize to new classes without forgetting previously learned classes. However, the semantic gap between old and new sample classes can lead to catastrophic forgetting. Additionally, existing algorithms lack capturing significant information from each sample image domain, impairing models' classification performance. Therefore, this paper proposes a novel Spatial-Frequency Domain Network (SFDNet) which contains a Spatial-Frequency Feature Extraction (SFFE) module and Attention Feature Alignment (AFA) module to improve the Zero-Shot Translation for Class Incremental algorithm. Firstly, SFFE module is designed which contains a dual attention mechanism for obtaining salient spatial-frequency feature information. Secondly, a novel feature fusion module is conducted for obtaining fused spatial-frequency domain features. Thirdly, the Nearest Class Mean classifier is utilized to select the most suitable category. Finally, iteration between tasks is performed using the Zero-Shot Translation model. The proposed SFDNet has the ability to effectively extract spatial-frequency feature representation from input images, improve the accuracy of image classification, and fundamentally alleviate catastrophic forgetting. Extensive experiments on the CUB 200-2011 and CIFAR100 datasets demonstrate that our proposed algorithm outperforms state-of-the-art incremental learning algorithms.