arXiv subjects

Haijin Zeng

Publications and source records attributed to Haijin Zeng.

At least 19 recordsLinked to original sources

arXiv2026

Predicting Functions, Not Features: KANs with Function-Space Joint-Embedding Predictive Learning for Medical Image Segmentation

Kolmogorov--Arnold Networks (KANs) introduce explicit functional representations by parameterizing each network edge as a learnable univariate function. However, existing KAN-based segmentation models optimize edge functions only through objectives defined after edge aggregation, leaving individual functions without an explicit pre-aggregation learning target. To address this limitation, we propose Function-Space Joint-Embedding Predictive Learning (FS-JEPA) for medical image segmentation. Our FS-JEPA framework moves predictive learning into the pre-aggregation function space of KANs. A masked online branch predicts structured signatures of sampled KAN edge functions generated by a full-context exponential moving average target branch, while shared edge indices preserve correspondence between predictions and targets. Rather than predicting an isolated edge response, we represent each sampled edge function using a multi-radius signature composed of function evaluations around its input anchor. This structured representation captures local functional variations that cannot be characterized by a single response and provides a more informative predictive target. The function-space objective is jointly optimized with the segmentation loss during training, while the predictive branch is removed at inference. Experiments on five medical image segmentation benchmarks show that our FS-JEPA achieves the best average Dice and outperforms the strongest competing KAN-based method by +2.25 percentage points.

cs.CV↗

arXiv2026

SegDem: Segmentation helps Demosaicing

Image demosaicing reconstructs a full-color image from incomplete color measurements produced by a sensor covered with a color filter array (CFA). Most existing methods formulate demosaicing as pixel-level reconstruction and mainly rely on local textures, cross-channel correlations, and low-level image statistics. Our core insight is that reconstruction and visual understanding can be viewed as complementary views of shared scene structure: both are grounded in the same underlying physical world, and therefore the structural and physical information inferred from an image should remain consistent across the two tasks. We instantiate this idea with instance segmentation and propose \emph{SegDem}, a cross-task decoder representation transfer framework for demosaicing. SegDem first learns region- and boundary-aware representations through instance-aware structural pretraining and then transfers the decoder to RAW-conditioned reconstruction. Segmentation- and demosaicing-conditioned features are further anchored to a shared frozen DINOv2 representation space to preserve structural organization across tasks. We instantiate SegDem with convolutional, Transformer-based, and state-space backbones for unified Single- and Quad-Bayer demosaicing. Extensive experiments on synthetic, external, and challenging datasets demonstrate consistent improvements across different architectures and CFA layouts.

cs.CV↗

arXiv2026

Structural Guidance for Unified Joint Demosaicing and Denoising

Joint demosaicing and denoising is a fundamental step in camera image signal processing, yet remains challenging because different Bayer-like color filter arrays (CFAs) and sensor noise jointly corrupt both color sampling and image content. Existing unified restoration networks explicitly model CFA geometry but are still driven primarily by pixel-level supervision, making them prone to structural degradation around edges, repetitive textures, and moir\'e patterns where local evidence is unreliable. We attribute this limitation partly to the absence of explicit structural guidance beyond pixel-level reconstruction supervision. Motivated by this observation, we propose a structural-guided unified restoration framework that injects pretrained structural knowledge into CFA-aware image restoration. Our model receives a unified five-channel observation consisting of the raw mosaic, CFA masks, and a noise-level map. A SwinIR restoration branch reconstructs pixel details under CFA-conditioned modulation, while a parallel structural reasoning branch extracts complementary structural cues from a sparse pseudo-RGB observation. To bridge the substantial domain gap between sparse noisy sensor data and the natural-image pretraining domain of the structural encoder, we introduce a lightweight trainable adapter before residually fusing structural and restoration features. A shared decoder jointly predicts the restored RGB image and an auxiliary clean mosaic, providing supervision in both image and sensor domains. Extensive experiments across multiple CFA patterns and noise levels demonstrate consistent improvements over state-of-the-art unified and CFA-specific methods, indicating that adapted structural priors can enhance robust camera image restoration. The source codes and dataset are provided in the supplementary material.

eess.IV↗

arXiv2026

The Second LoViF 2026 Challenge on Real-World All-in-One Image Restoration: Methods and Results

This paper presents a review of the second LoViF Challenge on Real-World All-in-One Image Restoration. The challenge aims to advance unified image restoration under diverse real-world degradation conditions, including blur, low-light, haze, rain, and snow. It provides a common benchmark for evaluating the restoration accuracy, robustness, and generalization capability of models across multiple degradation categories within a unified framework. The competition attracted 158 registered participants, and 20 teams were included in the final ranking after their submitted results were successfully reproduced and verified. This report provides a comprehensive analysis of the submitted solutions and corresponding results, highlighting recent advances in real-world all-in-one image restoration. The summarized methods and empirical findings reveal effective design strategies and establish an updated benchmark for future research in real-world low-level vision.

cs.CV↗

arXiv2026

Decoupling Language Guidance from Backbones for Text-Guided Medical Segmentation

Text-guided medical image segmentation leverages clinical semantics to improve lesion delineation, yet many existing models bind cross-modal fusion, supervision, and decoder design into a task-specific architecture. Such tight coupling makes it difficult to reuse language guidance modules across heterogeneous vision and text backbones, and often requires redesigning the network when the encoder pair changes. This paper presents BTHA, a backbone-transferable hierarchical adapter framework for text-guided medical image segmentation. BTHA is built around a stable feature-level interface: given multi-scale visual features and a text representation, it injects semantic guidance through shape-preserving adapters while maintaining the decoder-side tensor contract. To make this interface effective, we introduce a Hierarchical Coarse-to-Fine Supervision Strategy that decomposes learning into global image-text alignment, multi-scale auxiliary localization, and boundary-aware final mask refinement. We further design a Scale-Adaptive Gated Semantic Guidance (SAGSG) adapter, where resolution-specific gates adaptively control textual injection and channel recalibration suppresses redundant cross-modal responses. Evaluations across diverse vision and text backbones show that the same adapter and supervision design remains effective across convolutional and transformer-based visual encoders as well as different language encoders. Experiments on four public datasets further demonstrate that BTHA improves strong text-guided baselines with modest computational overhead.

cs.CV↗

arXiv2026

Differential Unfolding: Efficient Unfolding Reconstruction for Video Snapshot Compressive Imaging

While Deep Unfolding Networks (DUNs) dominate video Snapshot Compressive Imaging (SCI), they remain constrained by a uniform design philosophy. Existing methods repeatedly stack high-complexity priors with identical structures, ignoring the fact that optimization trajectories converge toward static states. This results in representation stagnation, where high-cost computations are wasted on minimal feature updates. To address this inefficiency, we present Differential Unfolding (DU), a heterogeneous framework that replaces uniform repetition with dynamic evolution. Central to DU is the Differential Evolutionary Framework (DEF), which partitions the unfolding process into two complementary roles: structural anchoring and differential evolution. In this scheme, high-parameter general stages are sparsely deployed to generate high-fidelity feature foundations. Complementing these, lightweight differential stages employ a Differential Representation Prior (DRP) to propagate and refine these foundational features through a differential mechanism. By integrating Differential Representation Attention (DRA) for evolving attention maps and a Differential Modulated FFN (DM-FFN) for feature rectification, DRP effectively models cross-stage variations with minimal overhead. By focusing computational resources on dynamic evolution rather than static redundancy, DU achieves a superior trade-off between accuracy and efficiency. Extensive experiments verify that our method establishes new state-of-the-art results while significantly slashing computational overhead. https://github.com/Muyuan-Zhang/DU

cs.CV↗

arXiv2026

Agentic Flow Steering and Parallel Rollout Search for Spatially Grounded Text-to-Image Generation

Precise Text-to-Image (T2I) generation has achieved great success but is hindered by the limited relational reasoning of static text encoders and the error accumulation in open-loop sampling. Without real-time feedback, initial semantic ambiguities during the Ordinary Differential Equation trajectory inevitably escalate into stochastic deviations from spatial constraints. To bridge this gap, we introduce AFS-Search (Agentic Flow Steering and Parallel Rollout Search), a training-free closed-loop framework built upon FLUX.1-dev. AFS-Search incorporates a training-free closed-loop parallel rollout search and flow steering mechanism, which leverages a Vision-Language Model (VLM) as a semantic critic to diagnose intermediate latents and dynamically steer the velocity field via precise spatial grounding. Complementarily, we formulate T2I generation as a sequential decision-making process, exploring multiple trajectories through lookahead simulations and selecting the optimal path based on VLM-guided rewards. Further, we provide AFS-Search-Pro for higher performance and AFS-Search-Fast for quicker generation. Experimental results show that our AFS-Search-Pro greatly boosts the performance of the original FLUX.1-dev, achieving state-of-the-art results across three different benchmarks. Meanwhile, AFS-Search-Fast also significantly enhances performance while maintaining fast generation speed.

cs.AI↗

arXiv2026

Deep LoRA-Unfolding Networks for Image Restoration

Deep unfolding networks (DUNs), combining conventional iterative optimization algorithms and deep neural networks into a multi-stage framework, have achieved remarkable accomplishments in Image Restoration (IR), such as spectral imaging reconstruction, compressive sensing and super-resolution.It unfolds the iterative optimization steps into a stack of sequentially linked blocks.Each block consists of a Gradient Descent Module (GDM) and a Proximal Mapping Module (PMM) which is equivalent to a denoiser from a Bayesian perspective, operating on Gaussian noise with a known level.However, existing DUNs suffer from two critical limitations: (i) their PMMs share identical architectures and denoising objectives across stages, ignoring the need for stage-specific adaptation to varying noise levels; and (ii) their chain of structurally repetitive blocks results in severe parameter redundancy and high memory consumption, hindering deployment in large-scale or resource-constrained scenarios.To address these challenges, we introduce generalized Deep Low-rank Adaptation (LoRA) Unfolding Networks for image restoration, named LoRun, harmonizing denoising objectives and adapting different denoising levels between stages with compressed memory usage for more efficient DUN.LoRun introduces a novel paradigm where a single pretrained base denoiser is shared across all stages, while lightweight, stage-specific LoRA adapters are injected into the PMMs to dynamically modulate denoising behavior according to the noise level at each unfolding step.This design decouples the core restoration capability from task-specific adaptation, enabling precise control over denoising intensity without duplicating full network parameters and achieving up to $N$ times parameter reduction for an $N$-stage DUN with on-par or better performance.Extensive experiments conducted on three IR tasks validate the efficiency of our method.

cs.CV↗

arXiv2026

HSI-VAR: Rethinking Hyperspectral Restoration through Spatial-Spectral Visual Autoregression

Hyperspectral images (HSIs) capture richer spatial-spectral information beyond RGB, yet real-world HSIs often suffer from a composite mix of degradations, such as noise, blur, and missing bands. Existing generative approaches for HSI restoration like diffusion models require hundreds of iterative steps, making them computationally impractical for high-dimensional HSIs. While regression models tend to produce oversmoothed results, failing to preserve critical structural details. We break this impasse by introducing HSI-VAR, rethinking HSI restoration as an autoregressive generation problem, where spectral and spatial dependencies can be progressively modeled rather than globally reconstructed. HSI-VAR incorporates three key innovations: (1) Latent-condition alignment, which couples semantic consistency between latent priors and conditional embeddings for precise reconstruction; (2) Degradation-aware guidance, which uniquely encodes mixed degradations as linear combinations in the embedding space for automatic control, remarkably achieving a nearly $50\%$ reduction in computational cost at inference; (3) A spatial-spectral adaptation module that refines details across both domains in the decoding phase. Extensive experiments on nine all-in-one HSI restoration benchmarks confirm HSI-VAR's state-of-the-art performance, achieving a 3.77 dB PSNR improvement on \textbf{\textit{ICVL}} and offering superior structure preservation with an inference speed-up of up to $95.5 \times$ compared with diffusion-based methods, making it a highly practical solution for real-world HSI restoration.

cs.CV↗

arXiv2026

Vision-Language Controlled Deep Unfolding for Joint Medical Image Restoration and Segmentation

We propose VL-DUN, a principled framework for joint All-in-One Medical Image Restoration and Segmentation (AiOMIRS) that bridges the gap between low-level signal recovery and high-level semantic understanding. While standard pipelines treat these tasks in isolation, our core insight is that they are fundamentally synergistic: restoration provides clean anatomical structures to improve segmentation, while semantic priors regularize the restoration process. VL-DUN resolves the sub-optimality of sequential processing through two primary innovations. (1) We formulate AiOMIRS as a unified optimization problem, deriving an interpretable joint unfolding mechanism where restoration and segmentation are mathematically coupled for mutual refinement. (2) We introduce a frequency-aware Mamba mechanism to capture long-range dependencies for global segmentation while preserving the high-frequency textures necessary for restoration. This allows for efficient global context modeling with linear complexity, effectively mitigating the spectral bias of standard architectures. As a pioneering work in the AiOMIRS task, VL-DUN establishes a new state-of-the-art across multi-modal benchmarks, improving PSNR by 0.92 dB and the Dice coefficient by 9.76\%. Our results demonstrate that joint collaborative learning offers a superior, more robust solution for complex clinical workflows compared to isolated task processing. The codes are provided in https://github.com/cipi666/VLDUN.

eess.IV↗

arXiv2025

BrainCognizer: Brain Decoding with Human Visual Cognition Simulation for fMRI-to-Image Reconstruction

Brain decoding is a key neuroscience field that reconstructs the visual stimuli from brain activity with fMRI, which helps illuminate how the brain represents the world. fMRI-to-image reconstruction has achieved impressive progress by leveraging diffusion models. However, brain signals infused with prior knowledge and associations exhibit a significant information asymmetry when compared to raw visual features, still posing challenges for decoding fMRI representations under the supervision of images. Consequently, the reconstructed images often lack fine-grained visual fidelity, such as missing attributes and distorted spatial relationships. To tackle this challenge, we propose BrainCognizer, a novel brain decoding model inspired by human visual cognition, which explores multi-level semantics and correlations without fine-tuning of generative models. Specifically, BrainCognizer introduces two modules: the Cognitive Integration Module which incorporates prior human knowledge to extract hierarchical region semantics; and the Cognitive Correlation Module which captures contextual semantic relationships across regions. Incorporating these two modules enhances intra-region semantic consistency and maintains inter-region contextual associations, thereby facilitating fine-grained brain decoding. Moreover, we quantitatively interpret our components from a neuroscience perspective and analyze the associations between different visual patterns and brain functions. Extensive quantitative and qualitative experiments demonstrate that BrainCognizer outperforms state-of-the-art approaches on multiple evaluation metrics.

q-bio.NC↗

arXiv2025

Gradient of White Matter Functional Variability via fALFF Differential Identifiability

Functional variability in both gray matter (GM) and white matter (WM) is closely associated with human brain cognitive and developmental processes, and is commonly assessed using functional connectivity (FC). However, as a correlation-based approach, FC captures the co-fluctuation between brain regions rather than the intensity of neural activity in each region. Consequently, FC provides only a partial view of functional variability, and this limitation is particularly pronounced in WM, where functional signals are weaker and more susceptible to noise. To tackle this limitation, we introduce fractional amplitude of low-frequency fluctuation (fALFF) to measure the intensity of spontaneous neural activity and analyze functional variability in WM. Specifically, we propose a novel method to quantify WM functional variability by estimating the differential identifiability of fALFF. Higher differential identifiability is observed in WM fALFF compared to FC, which indicates that fALFF is more sensitive to WM functional variability. Through fALFF differential identifiability, we evaluate the functional variabilities of both WM and GM, and find the overall functional variability pattern is similar although WM shows slightly lower variability than GM. The regional functional variabilities of WM are associated with structural connectivity, where commissural fiber regions generally exhibit higher variability than projection fiber regions. Furthermore, we discover that WM functional variability demonstrates a spatial gradient ascending from the brainstem to the cortex by hypothesis testing, which aligns well with the evolutionary expansion. The gradient of functional variability in WM provides novel insights for understanding WM function. To the best of our knowledge, this is the first attempt to investigate WM functional variability via fALFF.

q-bio.NC↗

arXiv2025

SlowFast-SCI: Slow-Fast Deep Unfolding Learning for Spectral Compressive Imaging

Humans learn in two complementary ways: a slow, cumulative process that builds broad, general knowledge, and a fast, on-the-fly process that captures specific experiences. Existing deep-unfolding methods for spectral compressive imaging (SCI) mirror only the slow component-relying on heavy pre-training with many unfolding stages-yet they lack the rapid adaptation needed to handle new optical configurations. As a result, they falter on out-of-distribution cameras, especially in bespoke spectral setups unseen during training. This depth also incurs heavy computation and slow inference. To bridge this gap, we introduce SlowFast-SCI, a dual-speed framework seamlessly integrated into any deep unfolding network beyond SCI systems. During slow learning, we pre-train or reuse a priors-based backbone and distill it via imaging guidance into a compact fast-unfolding model. In the fast learning stage, lightweight adaptation modules are embedded within each stage and fine-turned self-supervised at test time via a self-supervised loss-without retraining the backbone. To the best of our knowledge, SlowFast-SCI is the first testtime adaptation-driven deep unfolding framework for efficient, self-adaptive spectral reconstruction. Its dual-stage design unites offline robustness with on-the-fly per-sample calibration-yielding over 70% reduction in parameters and FLOPs, up to 5.79 dB PSNR improvement on out-of-distribution data, preserved cross-domain adaptability, and a 4x faster adaptation speed. In addition, its modularity integrates with any deep-unfolding network, paving the way for self-adaptive, field-deployable imaging and expanded computational imaging modalities. Code is available in Supplementary Material. The models, datasets, and code are available at https://github.com/XuanLu11/SlowFast-SCI.

cs.CV↗

arXiv2025

Lightweight Quad Bayer HybridEVS Demosaicing via State Space Augmented Cross-Attention

Event cameras like the Hybrid Event-based Vision Sensor (HybridEVS) camera capture brightness changes as asynchronous "events" instead of frames, offering advanced application on mobile photography. However, challenges arise from combining a Quad Bayer Color Filter Array (CFA) sensor with event pixels lacking color information, resulting in aliasing and artifacts on the demosaicing process before downstream application. Current methods struggle to address these issues, especially on resource-limited mobile devices. In response, we introduce \textbf{TSANet}, a lightweight \textbf{T}wo-stage network via \textbf{S}tate space augmented cross-\textbf{A}ttention, which can handle event pixels inpainting and demosaicing separately, leveraging the benefits of dividing complex tasks into manageable subtasks. Furthermore, we introduce a lightweight Cross-Swin State Block that uniquely utilizes positional prior for demosaicing and enhances global dependencies through the state space model with linear complexity. In summary, TSANet demonstrates excellent demosaicing performance on both simulated and real data of HybridEVS while maintaining a lightweight model, averaging better results than the previous state-of-the-art method DemosaicFormer across seven diverse datasets in both PSNR and SSIM, while respectively reducing parameter and computation costs by $1.86\times$ and $3.29\times$. Our approach presents new possibilities for efficient image demosaicing on mobile devices. Code is available in the supplementary materials.

cs.CV↗

arXiv2025

Vision-Language Gradient Descent-driven All-in-One Deep Unfolding Networks

Dynamic image degradations, including noise, blur and lighting inconsistencies, pose significant challenges in image restoration, often due to sensor limitations or adverse environmental conditions. Existing Deep Unfolding Networks (DUNs) offer stable restoration performance but require manual selection of degradation matrices for each degradation type, limiting their adaptability across diverse scenarios. To address this issue, we propose the Vision-Language-guided Unfolding Network (VLU-Net), a unified DUN framework for handling multiple degradation types simultaneously. VLU-Net leverages a Vision-Language Model (VLM) refined on degraded image-text pairs to align image features with degradation descriptions, selecting the appropriate transform for target degradation. By integrating an automatic VLM-based gradient estimation strategy into the Proximal Gradient Descent (PGD) algorithm, VLU-Net effectively tackles complex multi-degradation restoration tasks while maintaining interpretability. Furthermore, we design a hierarchical feature unfolding structure to enhance VLU-Net framework, efficiently synthesizing degradation patterns across various levels. VLU-Net is the first all-in-one DUN framework and outperforms current leading one-by-one and all-in-one end-to-end methods by 3.74 dB on the SOTS dehazing dataset and 1.70 dB on the Rain100L deraining dataset.

cs.CV↗

arXiv2025

Binarized Mamba-Transformer for Lightweight Quad Bayer HybridEVS Demosaicing

Quad Bayer demosaicing is the central challenge for enabling the widespread application of Hybrid Event-based Vision Sensors (HybridEVS). Although existing learning-based methods that leverage long-range dependency modeling have achieved promising results, their complexity severely limits deployment on mobile devices for real-world applications. To address these limitations, we propose a lightweight Mamba-based binary neural network designed for efficient and high-performing demosaicing of HybridEVS RAW images. First, to effectively capture both global and local dependencies, we introduce a hybrid Binarized Mamba-Transformer architecture that combines the strengths of the Mamba and Swin Transformer architectures. Next, to significantly reduce computational complexity, we propose a binarized Mamba (Bi-Mamba), which binarizes all projections while retaining the core Selective Scan in full precision. Bi-Mamba also incorporates additional global visual information to enhance global context and mitigate precision loss. We conduct quantitative and qualitative experiments to demonstrate the effectiveness of BMTNet in both performance and computational efficiency, providing a lightweight demosaicing solution suited for real-world edge devices. Our codes and models are available at https://github.com/Clausy9/BMTNet.

cs.CV↗

arXiv2025

Generative Model-Assisted Demosaicing for Cross-multispectral Cameras

As a crucial part of the spectral filter array (SFA)-based multispectral imaging process, spectral demosaicing has exploded with the proliferation of deep learning techniques. However, (1) bothering by the difficulty of capturing corresponding labels for real data or simulating the practical spectral imaging process, end-to-end networks trained in a supervised manner using simulated data often perform poorly on real data. (2) cross-camera spectral discrepancies make it difficult to apply pre-trained models to new cameras. (3) existing demosaicing networks are prone to introducing visual artifacts on hard cases due to the interpolation of unknown values. To address these issues, we propose a hybrid supervised training method with the assistance of the self-supervised generative model, which performs well on real data across different spectral cameras. Specifically, our approach consists of three steps: (1) Pre-Training step: training the end-to-end neural network on a large amount of simulated data; (2) Pseudo-Pairing step: generating pseudo-labels of real target data using the self-supervised generative model; (3) Fine-Tuning step: fine-tuning the pre-trained model on the pseudo data pairs obtained in (2). To alleviate artifacts, we propose a frequency-domain hard patch selection method that identifies artifact-prone regions by analyzing spectral discrepancies using Fourier transform and filtering techniques, allowing targeted fine-tuning to enhance demosaicing performance. Finally, we propose UniSpecTest, a real-world multispectral mosaic image dataset for testing. Ablation experiments have demonstrated the effectiveness of each training step, and extensive experiments on both synthetic and real datasets show that our method achieves significant performance gains compared to state-of-the-art techniques.

eess.IV↗

arXiv2024

OTLRM: Orthogonal Learning-based Low-Rank Metric for Multi-Dimensional Inverse Problems

In real-world scenarios, complex data such as multispectral images and multi-frame videos inherently exhibit robust low-rank property. This property is vital for multi-dimensional inverse problems, such as tensor completion, spectral imaging reconstruction, and multispectral image denoising. Existing tensor singular value decomposition (t-SVD) definitions rely on hand-designed or pre-given transforms, which lack flexibility for defining tensor nuclear norm (TNN). The TNN-regularized optimization problem is solved by the singular value thresholding (SVT) operator, which leverages the t-SVD framework to obtain the low-rank tensor. However, it is quite complicated to introduce SVT into deep neural networks due to the numerical instability problem in solving the derivatives of the eigenvectors. In this paper, we introduce a novel data-driven generative low-rank t-SVD model based on the learnable orthogonal transform, which can be naturally solved under its representation. Prompted by the linear algebra theorem of the Householder transformation, our learnable orthogonal transform is achieved by constructing an endogenously orthogonal matrix adaptable to neural networks, optimizing it as arbitrary orthogonal matrices. Additionally, we propose a low-rank solver as a generalization of SVT, which utilizes an efficient representation of generative networks to obtain low-rank structures. Extensive experiments highlight its significant restoration enhancements.

cs.CV↗