arXiv subjects

Zhao Zhang

Publications and source records attributed to Zhao Zhang.

At least 145 records · Page 8Linked to original sources

arXiv2024

Sagiri: Low Dynamic Range Image Enhancement with Generative Diffusion Prior

Capturing High Dynamic Range (HDR) scenery using 8-bit cameras often suffers from over-/underexposure, loss of fine details due to low bit-depth compression, skewed color distributions, and strong noise in dark areas. Traditional LDR image enhancement methods primarily focus on color mapping, which enhances the visual representation by expanding the image's color range and adjusting the brightness. However, these approaches fail to effectively restore content in dynamic range extremes, which are regions with pixel values close to 0 or 255. To address the full scope of challenges in HDR imaging and surpass the limitations of current models, we propose a novel two-stage approach. The first stage maps the color and brightness to an appropriate range while keeping the existing details, and the second stage utilizes a diffusion prior to generate content in dynamic range extremes lost during capture. This generative refinement module can also be used as a plug-and-play module to enhance and complement existing LDR enhancement models. The proposed method markedly improves the quality and details of LDR images, demonstrating superior performance through rigorous experimental validation. The project page is at https://sagiri0208.github.io

eess.IV↗

arXiv2024

MIPI 2024 Challenge on Nighttime Flare Removal: Methods and Results

The increasing demand for computational photography and imaging on mobile platforms has led to the widespread development and integration of advanced image sensors with novel algorithms in camera systems. However, the scarcity of high-quality data for research and the rare opportunity for in-depth exchange of views from industry and academia constrain the development of mobile intelligent photography and imaging (MIPI). Building on the achievements of the previous MIPI Workshops held at ECCV 2022 and CVPR 2023, we introduce our third MIPI challenge including three tracks focusing on novel image sensors and imaging algorithms. In this paper, we summarize and review the Nighttime Flare Removal track on MIPI 2024. In total, 170 participants were successfully registered, and 14 teams submitted results in the final testing phase. The developed solutions in this challenge achieved state-of-the-art performance on Nighttime Flare Removal. More details of this challenge and the link to the dataset can be found at https://mipi-challenge.org/MIPI2024/.

cs.CV↗

arXiv2024

DFGNN: Dual-frequency Graph Neural Network for Sign-aware Feedback

The graph-based recommendation has achieved great success in recent years. However, most existing graph-based recommendations focus on capturing user preference based on positive edges/feedback, while ignoring negative edges/feedback (e.g., dislike, low rating) that widely exist in real-world recommender systems. How to utilize negative feedback in graph-based recommendations still remains underexplored. In this study, we first conducted a comprehensive experimental analysis and found that (1) existing graph neural networks are not well-suited for modeling negative feedback, which acts as a high-frequency signal in a user-item graph. (2) The graph-based recommendation suffers from the representation degeneration problem. Based on the two observations, we propose a novel model that models positive and negative feedback from a frequency filter perspective called Dual-frequency Graph Neural Network for Sign-aware Recommendation (DFGNN). Specifically, in DFGNN, the designed dual-frequency graph filter (DGF) captures both low-frequency and high-frequency signals that contain positive and negative feedback. Furthermore, the proposed signed graph regularization is applied to maintain the user/item embedding uniform in the embedding space to alleviate the representation degeneration problem. Additionally, we conduct extensive experiments on real-world datasets and demonstrate the effectiveness of the proposed model. Codes of our model will be released upon acceptance.

cs.IR↗

arXiv2024

Out of Many, One: Designing and Scaffolding Proteins at the Scale of the Structural Universe with Genie 2

Protein diffusion models have emerged as a promising approach for protein design. One such pioneering model is Genie, a method that asymmetrically represents protein structures during the forward and backward processes, using simple Gaussian noising for the former and expressive SE(3)-equivariant attention for the latter. In this work we introduce Genie 2, extending Genie to capture a larger and more diverse protein structure space through architectural innovations and massive data augmentation. Genie 2 adds motif scaffolding capabilities via a novel multi-motif framework that designs co-occurring motifs with unspecified inter-motif positions and orientations. This makes possible complex protein designs that engage multiple interaction partners and perform multiple functions. On both unconditional and conditional generation, Genie 2 achieves state-of-the-art performance, outperforming all known methods on key design metrics including designability, diversity, and novelty. Genie 2 also solves more motif scaffolding problems than other methods and does so with more unique and varied solutions. Taken together, these advances set a new standard for structure-based protein design. Genie 2 inference and training code, as well as model weights, are freely available at: https://github.com/aqlaboratory/genie2.

q-bio.BM↗

arXiv2024

Cross-Domain Knowledge Distillation for Low-Resolution Human Pose Estimation

In practical applications of human pose estimation, low-resolution inputs frequently occur, and existing state-of-the-art models perform poorly with low-resolution images. This work focuses on boosting the performance of low-resolution models by distilling knowledge from a high-resolution model. However, we face the challenge of feature size mismatch and class number mismatch when applying knowledge distillation to networks with different input resolutions. To address this issue, we propose a novel cross-domain knowledge distillation (CDKD) framework. In this framework, we construct a scale-adaptive projector ensemble (SAPE) module to spatially align feature maps between models of varying input resolutions. It adopts a projector ensemble to map low-resolution features into multiple common spaces and adaptively merges them based on multi-scale information to match high-resolution features. Additionally, we construct a cross-class alignment (CCA) module to solve the problem of the mismatch of class numbers. By combining an easy-to-hard training (ETHT) strategy, the CCA module further enhances the distillation performance. The effectiveness and efficiency of our approach are demonstrated by extensive experiments on two common benchmark datasets: MPII and COCO. The code is made available in supplementary material.

cs.CV↗

arXiv2024

Focus on Low-Resolution Information: Multi-Granular Information-Lossless Model for Low-Resolution Human Pose Estimation

In real-world applications of human pose estimation, low-resolution input images are frequently encountered when the performance of the image acquisition equipment is limited or the shooting distance is too far. However, existing state-of-the-art models for human pose estimation perform poorly on low-resolution images. One key reason is the presence of downsampling layers in these models, e.g., strided convolutions and pooling layers. It further reduces the already insufficient image information. Another key reason is that the body skeleton and human kinematic information are not fully utilized. In this work, we propose a Multi-Granular Information-Lossless (MGIL) model to replace the downsampling layers to address the above issues. Specifically, MGIL employs a Fine-grained Lossless Information Extraction (FLIE) module, which can prevent the loss of local information. Furthermore, we design a Coarse-grained Information Interaction (CII) module to adequately leverage human body structural information. To efficiently fuse cross-granular information and thoroughly exploit the relationships among keypoints, we further introduce a Multi-Granular Adaptive Fusion (MGAF) mechanism. The mechanism assigns weights to features of different granularities based on the content of the image. The model is effective, flexible, and universal. We show its potential in various vision tasks with comprehensive experiments. It outperforms the SOTA methods by 7.7 mAP on COCO and performs well with different input resolutions, different backbones, and different vision tasks. The code is provided in supplementary material.

cs.CV↗

arXiv2024

GinAR: An End-To-End Multivariate Time Series Forecasting Model Suitable for Variable Missing

Multivariate time series forecasting (MTSF) is crucial for decision-making to precisely forecast the future values/trends, based on the complex relationships identified from historical observations of multiple sequences. Recently, Spatial-Temporal Graph Neural Networks (STGNNs) have gradually become the theme of MTSF model as their powerful capability in mining spatial-temporal dependencies, but almost of them heavily rely on the assumption of historical data integrity. In reality, due to factors such as data collector failures and time-consuming repairment, it is extremely challenging to collect the whole historical observations without missing any variable. In this case, STGNNs can only utilize a subset of normal variables and easily suffer from the incorrect spatial-temporal dependency modeling issue, resulting in the degradation of their forecasting performance. To address the problem, in this paper, we propose a novel Graph Interpolation Attention Recursive Network (named GinAR) to precisely model the spatial-temporal dependencies over the limited collected data for forecasting. In GinAR, it consists of two key components, that is, interpolation attention and adaptive graph convolution to take place of the fully connected layer of simple recursive units, and thus are capable of recovering all missing variables and reconstructing the correct spatial-temporal dependencies for recursively modeling of multivariate time series data, respectively. Extensive experiments conducted on five real-world datasets demonstrate that GinAR outperforms 11 SOTA baselines, and even when 90% of variables are missing, it can still accurately predict the future values of all variables.

cs.LG↗

arXiv2024

ID-centric Pre-training for Recommendation

Classical sequential recommendation models generally adopt ID embeddings to store knowledge learned from user historical behaviors and represent items. However, these unique IDs are challenging to be transferred to new domains. With the thriving of pre-trained language model (PLM), some pioneer works adopt PLM for pre-trained recommendation, where modality information (e.g., text) is considered universal across domains via PLM. Unfortunately, the behavioral information in ID embeddings is still verified to be dominating in PLM-based recommendation models compared to modality information and thus limits these models' performance. In this work, we propose a novel ID-centric recommendation pre-training paradigm (IDP), which directly transfers informative ID embeddings learned in pre-training domains to item representations in new domains. Specifically, in pre-training stage, besides the ID-based sequential model for recommendation, we also build a Cross-domain ID-matcher (CDIM) learned by both behavioral and modality information. In the tuning stage, modality information of new domain items is regarded as a cross-domain bridge built by CDIM. We first leverage the textual information of downstream domain items to retrieve behaviorally and semantically similar items from pre-training domains using CDIM. Next, these retrieved pre-trained ID embeddings, rather than certain textual embeddings, are directly adopted to generate downstream new items' embeddings. Through extensive experiments on real-world datasets, both in cold and warm settings, we demonstrate that our proposed model significantly outperforms all baselines. Codes will be released upon acceptance.

cs.IR↗

arXiv2024

Validating a lutetium frequency reference

We review our progress in developing a frequency reference with singly ionized lutetium and give estimates of the levels of inaccuracy we expect to achieve in the near future with both the $^1S_0\leftrightarrow{}^3D_1$ and $^1S_0\leftrightarrow{}^3D_2$ transitions. Based on established experimental results, we show that inaccuracies at the low $10^{-19}$ level are readily achievable for the $^1S_0\leftrightarrow{}^3D_1$ transition, and the frequency ratio between the two transitions is limited almost entirely by the BBR shift. We argue that the frequency ratio measured within the one apparatus provides a well-defined metric to compare and establish the performance of remotely located systems. For the measurement of an in situ frequency ratio, relativistic shifts drop out and both transitions experience the same electromagnetic environment. Consequently, the uncertainty budget for the ratio is practically identical to the uncertainty budgets for the individual transitions. If the ratios for two or more systems disagree we can be certain at least one of the clock assessments is incorrect. If they agree, subsequent comparisons on one transition would only differ by relativistic effects. Since motional effects are easily assessed and typically small for a heavy ion, only the differential gravitational red-shift will significantly contribute and this can be confirmed by comparison on the second transition.

physics.atom-ph↗

arXiv2024

NTIRE 2024 Challenge on Low Light Image Enhancement: Methods and Results

This paper reviews the NTIRE 2024 low light image enhancement challenge, highlighting the proposed solutions and results. The aim of this challenge is to discover an effective network design or solution capable of generating brighter, clearer, and visually appealing results when dealing with a variety of conditions, including ultra-high resolution (4K and beyond), non-uniform illumination, backlighting, extreme darkness, and night scenes. A notable total of 428 participants registered for the challenge, with 22 teams ultimately making valid submissions. This paper meticulously evaluates the state-of-the-art advancements in enhancing low-light images, reflecting the significant progress and creativity in this field.

cs.CV↗

arXiv2024

Graphic Design with Large Multimodal Model

In the field of graphic design, automating the integration of design elements into a cohesive multi-layered artwork not only boosts productivity but also paves the way for the democratization of graphic design. One existing practice is Graphic Layout Generation (GLG), which aims to layout sequential design elements. It has been constrained by the necessity for a predefined correct sequence of layers, thus limiting creative potential and increasing user workload. In this paper, we present Hierarchical Layout Generation (HLG) as a more flexible and pragmatic setup, which creates graphic composition from unordered sets of design elements. To tackle the HLG task, we introduce Graphist, the first layout generation model based on large multimodal models. Graphist efficiently reframes the HLG as a sequence generation problem, utilizing RGB-A images as input, outputs a JSON draft protocol, indicating the coordinates, size, and order of each element. We develop new evaluation metrics for HLG. Graphist outperforms prior arts and establishes a strong baseline for this field. Project homepage: https://github.com/graphic-design-ai/graphist

cs.CV↗

arXiv2024

MV-Adapter: Multimodal Video Transfer Learning for Video Text Retrieval

State-of-the-art video-text retrieval (VTR) methods typically involve fully fine-tuning a pre-trained model (e.g. CLIP) on specific datasets. However, this can result in significant storage costs in practical applications as a separate model per task must be stored. To address this issue, we present our pioneering work that enables parameter-efficient VTR using a pre-trained model, with only a small number of tunable parameters during training. Towards this goal, we propose a new method dubbed Multimodal Video Adapter (MV-Adapter) for efficiently transferring the knowledge in the pre-trained CLIP from image-text to video-text. Specifically, MV-Adapter utilizes bottleneck structures in both video and text branches, along with two novel components. The first is a Temporal Adaptation Module that is incorporated in the video branch to introduce global and local temporal contexts. We also train weights calibrations to adjust to dynamic variations across frames. The second is Cross Modality Tying that generates weights for video/text branches through sharing cross modality factors, for better aligning between modalities. Thanks to above innovations, MV-Adapter can achieve comparable or better performance than standard full fine-tuning with negligible parameters overhead. Notably, MV-Adapter consistently outperforms various competing methods in V2T/T2V tasks with large margins on five widely used VTR benchmarks (MSR-VTT, MSVD, LSMDC, DiDemo, and ActivityNet).

cs.CV↗

arXiv2024

Multimodal Road Network Generation Based on Large Language Model

With the increasing popularity of ChatGPT, large language models (LLMs) have demonstrated their capabilities in communication and reasoning, promising for transportation sector intelligentization. However, they still face challenges in domain-specific knowledge. This paper aims to leverage LLMs' reasoning and recognition abilities to replace traditional user interfaces and create an "intelligent operating system" for transportation simulation software, exploring their potential with transportation modeling and simulation. We introduce Network Generation AI (NGAI), integrating LLMs with road network modeling plugins, validated through experiments for accuracy and robustness. NGAI's effective use has reduced modeling costs, revolutionized transportation simulations, optimized user steps, and proposed a novel approach for LLM integration in the transportation field.

cs.HC↗

arXiv2024

Dark matter phenomenology and phase transition dynamics of the next to minimal composite Higgs model with dilaton

In this paper, we conduct a comprehensive study of the Next-to-Minimal Composite Higgs Model (NMCHM) extended with a dilaton field $χ$ (denoted as NMCHM$_χ$). A pseudo-Nambu-Goldstone boson (pNGB) $η$, resulting from the SO(6)$\to$SO(5) breaking, serves as a dark matter (DM) candidate. The inclusion of the dilaton field is helpful for evading the stringent constraints from dark matter direct detection, as it allows for an accidental cancellation between the amplitudes of DM-nucleon scattering, an outcome of the mixing between the dilaton and Higgs fields. The presence of the dilaton field also enriches the phase transition patterns in the early universe. We identify two types of phase transitions: (i) a 1-step phase transition, where the chiral symmetry and electroweak symmetry breaking (EWSB) occur simultaneously, and (ii) a 2-step phase transition, where the chiral symmetry breaking transition takes place first, followed by a second phase transition corresponding to EWSB. Since the first-order phase transitions can be strong due to supercooling in our model, we also examine the stochastic background of gravitational waves generated by these phase transitions. We find that these gravitational waves hold promise for detection in future space-based gravitational wave experiments, such as LISA, Taiji, BBO, and DECIGO.

hep-ph↗

arXiv2024

Self-Improvement Programming for Temporal Knowledge Graph Question Answering

Temporal Knowledge Graph Question Answering (TKGQA) aims to answer questions with temporal intent over Temporal Knowledge Graphs (TKGs). The core challenge of this task lies in understanding the complex semantic information regarding multiple types of time constraints (e.g., before, first) in questions. Existing end-to-end methods implicitly model the time constraints by learning time-aware embeddings of questions and candidate answers, which is far from understanding the question comprehensively. Motivated by semantic-parsing-based approaches that explicitly model constraints in questions by generating logical forms with symbolic operators, we design fundamental temporal operators for time constraints and introduce a novel self-improvement Programming method for TKGQA (Prog-TQA). Specifically, Prog-TQA leverages the in-context learning ability of Large Language Models (LLMs) to understand the combinatory time constraints in the questions and generate corresponding program drafts with a few examples given. Then, it aligns these drafts to TKGs with the linking module and subsequently executes them to generate the answers. To enhance the ability to understand questions, Prog-TQA is further equipped with a self-improvement strategy to effectively bootstrap LLMs using high-quality self-generated drafts. Extensive experiments demonstrate the superiority of the proposed Prog-TQA on MultiTQ and CronQuestions datasets, especially in the Hits@1 metric.

cs.CL↗

arXiv2024

GT-Rain Single Image Deraining Challenge Report

This report reviews the results of the GT-Rain challenge on single image deraining at the UG2+ workshop at CVPR 2023. The aim of this competition is to study the rainy weather phenomenon in real world scenarios, provide a novel real world rainy image dataset, and to spark innovative ideas that will further the development of single image deraining methods on real images. Submissions were trained on the GT-Rain dataset and evaluated on an extension of the dataset consisting of 15 additional scenes. Scenes in GT-Rain are comprised of real rainy image and ground truth image captured moments after the rain had stopped. 275 participants were registered in the challenge and 55 competed in the final testing phase.

cs.CV↗

arXiv2024

Enhanced micromotion compensation using a phase modulated light field

We investigate sideband spectroscopy of a trapped ion using a probe laser phase modulated at the trap drive frequency. The enhanced sensitivity of our technique over traditional sideband spectroscopy allows us to detect stray fields of $0.01\,\mathrm{V/m}$ on a timescale of a few minutes and detect differential phases of $5\,μ\mathrm{rad}$ between applied ac potentials. We also demonstrate the ability suppress Doppler shifts from excess motion to well below the limit imposed by the intrinsic motion of the ion in the vibrational ground-state. The technique we introduce can be readily implemented in any ion trap system that utilizes sideband spectroscopy for micromotion compensation and can be seamlessly integrated into experiments in a fully automated way

physics.atom-ph↗

arXiv2024

The First Place Solution of WSDM Cup 2024: Leveraging Large Language Models for Conversational Multi-Doc QA

Conversational multi-doc question answering aims to answer specific questions based on the retrieved documents as well as the contextual conversations. In this paper, we introduce our winning approach for the "Conversational Multi-Doc QA" challenge in WSDM Cup 2024, which exploits the superior natural language understanding and generation capability of Large Language Models (LLMs). We first adapt LLMs to the task, then devise a hybrid training strategy to make the most of in-domain unlabeled data. Moreover, an advanced text embedding model is adopted to filter out potentially irrelevant documents and several approaches are designed and compared for the model ensemble. Equipped with all these techniques, our solution finally ranked 1st place in WSDM Cup 2024, surpassing its rivals to a large extent. The source codes have been released at https://github.com/zhangzhao219/WSDM-Cup-2024.

cs.CL↗