Multi-arm collaboration is becoming a core capability in embodied manipulation. Recent vision-language-action (VLA) models integrate perception, language, and control, but most represent language as a single global instruction and do not provide an explicit mechanism for assigning and composing arm-specific behaviors. This design limits transfer to collaboration patterns that differ from those observed during training. We present MA-VLA, a unified framework for multi-arm collaboration via atomic action assignment. MA-VLA decomposes cooperative behavior into mid-level atomic prompts and allocates them to individual arms, enabling explicit subgoal specification and compositional reuse across tasks. To reduce reliance on fixed execution roles, we introduce Arm Shuffle, a training-time permutation of the observation, state, and assigned atomic prompts for each arm. This permutation enforces role-agnostic instruction following and supports recomposition into unseen coordination patterns, which we term multi-arm compositional generalization. We also construct a benchmark in which test-time collaboration patterns are absent in training set. Across simulation and real-world evaluations, prior state-of-the-art VLAs largely fail under these unseen collaborations, while MA-VLA consistently succeeds. These results indicate that structured, per-arm atomic action assignment offers a practical route to scalable generalization in multi-arm embodied systems. Code, models, and data are available at https://github.com/zhangzaibin/future-robots

This paper presents the NTIRE 2026 Remote Sensing Infrared Image Super-Resolution (x4) Challenge, one of the associated challenges of NTIRE 2026. The challenge aims to recover high-resolution (HR) infrared images from low-resolution (LR) inputs generated through bicubic downsampling with a x4 scaling factor. The objective is to develop effective models or solutions that achieve state-of-the-art performance for infrared image SR in remote sensing scenarios. To reflect the characteristics of infrared data and practical application needs, the challenge adopts a single-track setting. A total of 115 participants registered for the competition, with 13 teams submitting valid entries. This report summarizes the challenge design, dataset, evaluation protocol, main results, and the representative methods of each team. The challenge serves as a benchmark to advance research in infrared image super-resolution and promote the development of effective solutions for real-world remote sensing applications.

Depth estimation is a fundamental task in 3D computer vision, crucial for applications such as 3D reconstruction, free-viewpoint rendering, robotics, autonomous driving, and AR/VR technologies. Traditional methods relying on hardware sensors like LiDAR are often limited by their high costs, low resolution, and sensitivity to the environment, limiting their applicability to real-world scenarios. Recent advances in vision-based methods offer a promising alternative, yet they face challenges in generalization and stability due to either the low capacity of model architectures or reliance on domain-specific and small-scale datasets. The emergence of scaling laws and foundation models in other domains has inspired the development of “depth foundation models”: deep neural networks trained on large datasets with strong zeroshot generalization capabilities. This paper surveys the evolution of deep learning architectures and paradigms for depth estimation across monocular, stereo, multiview, and monocular video settings. We explore the potential of these models to address existing challenges and we also provide a comprehensive overview of large-scale datasets that can facilitate their development. By identifying key architectures and training strategies, we aim to highlight the path towards robust depth foundation models, offering insights for future research and applications.

Photorealistic Style Transfer (PST) aims to transfer the color and tonal style of a reference to a content image while strictly preserving its structural integrity. However, existing deep learning-based methods inherently suffer from semantic entanglement caused by pre-trained image encoders, leading to unnatural spatial distortions. Moreover, current pixel-level mapping paradigms often ignore color gamut topology, resulting in color banding, while also lacking the multimodal capability for intuitive text-driven control. To address these bottlenecks, we propose StatLUT, an innovative multimodal framework for 3D LUT generation. First, we bypass traditional encoders and introduce a Lab-Extractor to derive spatially-agnostic statistical features, fundamentally decoupling color distributions from structural semantics to ensure artifact-free rendering. Second, we formulate LUT generation as a Transformer-based Seq2Seq translation task, utilizing a Multi-dimensional Residual Mapper (MR-Mapper) to predict topologically smooth 3D LUTs. Finally, to break the single-modal barrier, we propose the H-Diffuser, a lightweight Diffusion Transformer that directly synthesizes statistical features from natural language prompts, enabling flexible text-driven color grading. Extensive experiments on standard benchmarks demonstrate that StatLUT significantly outperforms state-of-the-art methods in both visual quality and quantitative metrics, pioneering a highly robust and flexible paradigm for multimodal photorealistic style transfer.

The rapid advancement of image generation and editing techniques has rendered the detection and precise localization of forged content an increasingly demanding task. Existing approaches predominantly rely on learning forgery-specific artifacts, which limits their ability to generalize to unseen manipulation types. In this work, we introduce a Generalizable Image Forgery Localization (GIFL) framework that reframes the problem: rather than seeking manipulation traces, we propose to model the intrinsic distribution of authentic image content. GIFL learns a universal, content-consistent representation from pristine regions, organizes the feature space to naturally separate manipulated areas, and constructs a cohesive representation of authenticity that generalizes across diverse forgery types. To further improve robustness, we design a dual-domain interaction module that integrates complementary spectral and spatial cues for reliable localization. Additionally, to advance research on forgeries produced by modern deep generative models, we present Forgery ADE, a new comprehensive dataset containing images edited with a variety of popular deep image editing methods. Extensive experiments demonstrate that our method outperforms existing methods in localizing unseen forgeries also demonstrates competitive results on trained manipulation types, offering a more practical and robust solution for image authenticity verification in the era of generative AI.

In the absence of decoding specifications provided by the original equipment manufacturer (OEM), CAN protocol reverse engineering infers the signal structure by analyzing communication data on the CAN bus network. Existing CAN reverse engineering approaches can generally be categorized into two types: combinatorial optimization–based methods and bit-flip rate–based methods. Combinatorial optimization–based methods explore possible signal partitioning schemes and determine signal boundaries by comparing them with OBD or sensor data. However, these methods depend on external reference data, offer limited coverage of signal types, and consequently struggle to comprehensively identify all signal fields within a data frame. Bit-flip rate–based methods calculate the flip rate of each bit and identify abrupt change points to segment signal boundaries; however, these methods often suffer from a high false positive rate. Since CAN signals are expected to exhibit temporal smoothness, incorrect segmentation can cause significant fluctuations in the decoded results and degrade overall smoothness. In this study, quantified smoothness is used as an evaluation metric. Building upon byte-level segmentation, various boundary schemes are assessed by comparing the smoothness of the resulting signals, thereby determining the accuracy of signal boundaries. This approach prevents misjudgments that can arise from relying solely on abrupt changes in bit-flip rates for boundary detection.

Deep image inpainting techniques produce visually seamless forgeries that pose unique challenges to conventional forgery detectors due to their local coherence and semantic consistency. To address this, we propose a Fine-Grained Graph Convolution Network (IFL-GCN) for Inpainting Forgery Localization. Departing from prior local-global forgery detectors, IFL-GCN introduces a pixel-wise graph construction that achieves direct integration of local forgery traces across the entire image. By modeling each pixel as a graph node, our approach captures long-range inconsistencies while preserving fine-grained detail perception, enabling comprehensive analysis of irregular artifact distributions at the global scale. Additionally, we develop a Fidelity-aware Weighted Loss (FW loss) to dynamically calibrate learning objectives based on the estimated fidelity of the forged content, enhancing the detector's sensitivity to subtle, high-fidelity forgeries. Furthermore, to bridge the generalization gap across diverse inpainting artifacts and improve robustness, we propose Forgery Intensity Mixup augmentation (FIM), which expands the distribution of forgery features by modulating the intensity of forgeries while maintaining spatial-semantic integrity. Extensive experiments across 6 mainstream forgery detection benchmarks demonstrate that IFL-GCN achieves state-of-the-art performance, outperforming the closest competing method by 6.7% in average F1 score over all inpainting forgery test sets.

Deep learning has reshaped vascular biometrics from a recognition-centered task into a broader research field involving image refinement, vascular representation learning, synthetic data generation, and security and privacy protection. This shift is important because vascular traits encode identity through subtle, structure-rich, and modality-dependent vascular patterns, while practical systems still face challenges related to data scarcity, heterogeneous acquisition devices, inconsistent evaluation protocols, cross-domain degradation, and privacy risks. This paper provides a comprehensive survey of vascular biometrics in the deep learning era. We first review representative vascular modalities, including finger, palm, dorsal hand, and wrist veins, as well as retinal and scleral vessels, together with their acquisition characteristics, datasets, evaluation protocols, and metrics. We then propose a new taxonomy that reorganizes existing deep learning studies into four key dimensions: vascular image refinement, vascular representation learning, synthetic vascular generation, and security and privacy. Based on this taxonomy, we analyze representative methods, summarize recent technical progress, and review applications in identity authentication, health monitoring, and disease detection. Finally, we discuss open challenges and future directions. This survey aims to provide a systematic reference for understanding the methodological evolution, current landscape, and future development of deep vascular biometrics.

While Vision-Language Models (VLMs) excel at 2D visual understanding, they remain constrained by 2D-centric paradigm that severely limits genuine 3D spatial reasoning. To bridge this gap, we introduce Think3D, a novel framework that equips VLM agents with interactive, 3D chain-of-thought reasoning capabilities. By integrating a suite of 3D manipulation tools, Think3D transforms perception into active spatial exploration, mirroring human geometric reasoning. Think3D consistently improves proprietary models, including GPT-4.1 and Gemini 2.5 Pro, across BLINK Multi-view, MindCube-1K, and VSI-Bench-Tiny. We further propose Think3D-RL to teach smaller open-weight models how to manipulate 3D space effectively. Using only final-answer rewards, without process supervision or handcrafted exploration trajectories, Think3D-RL enables Qwen3-VL-4B to autonomously learn effective 3D exploration strategies. After training, the model exhibits tool-use patterns similar to those of stronger proprietary models, while shifting the effect of 3D tool use on MindCube-1K from a performance drop to a substantial improvement. These results show that active exploration in 3D space provides an effective and general paradigm for improving spatial reasoning in multimodal agents. Code, models, and data are available at https://github.com/zhangzaibin/spagent.

Monocular 3D object detection has received considerable attention for its simplicity and low cost. Existing methods typically follow conventional 2D detection paradigms, first locating object centers and then predicting 3D attributes via neighboring features. However, these approaches mainly focus on local information, which may limit the model's global context awareness and result in missed detections, as the global context provides semantic and spatial dependencies essential for detecting small objects in cluttered or occluded environments. In addition, due to large variation in object scales across different scenes and depths, inaccurate receptive fields often lead to background noise and degraded feature representation. To address these issues, we introduce MonoASRH, a novel monocular 3D detection framework composed of Efficient Hybrid Feature Aggregation Module (EH-FAM) and Adaptive Scale-Aware 3D Regression Head (ASRH). Specifically, EH-FAM employs multi-head attention with a global receptive field to extract semantic features and leverages lightweight convolutional modules to efficiently aggregate visual features across different scales, enhancing small-scale object detection. The ASRH encodes 2D bounding box dimensions and then fuses scale features with the semantic features aggregated by EH-FAM through a scale-semantic feature fusion module. The scale-semantic feature fusion module guides ASRH in learning dynamic receptive field offsets, incorporating scale information into 3D position prediction for better scale-awareness. Extensive experiments on the KITTI and Waymo datasets demonstrate that MonoASRH achieves state-of-the-art performance. The code and model are released at https://github.com/WYFDUT/MonoASRH

Salient object detection is inherently a subjective problem, as observers with different priors may perceive different objects as salient. However, existing methods predominantly formulate it as an objective prediction task with a single groundtruth segmentation map for each image, which renders the problem under-determined and fundamentally ill-posed. To address this issue, we propose Observer-Centric Salient Object Detection (OC-SOD), where salient regions are predicted by considering not only the visual cues but also the observer-specific factors such as their preferences or intents. As a result, this formulation captures the intrinsic ambiguity and diversity of human perception, enabling personalized and context-aware saliency prediction. By leveraging multi-modal large language models, we develop an efficient data annotation pipeline and construct the first OC-SOD dataset named OC-SODBench, comprising 33k training, validation and test images with 152k textual prompts and object pairs. Built upon this new dataset, we further design OC-SODAgent, an agentic baseline which performs OC-SOD via a human-like "Perceive-Reflect-Adjust" process. Extensive experiments on our proposed OC-SODBench have justified the effectiveness of our contribution. Through this observer-centric perspective, we aim to bridge the gap between human perception and computational modeling, offering a more realistic and flexible understanding of what makes an object truly "salient." Code and dataset are publicly available at: https://github.com/Dustzx/OC_SOD

We introduce Illustrator's Depth, a novel definition of depth that addresses a key challenge in digital content creation: decomposing flat images into editable, ordered layers. Inspired by an artist's compositional process, illustrator's depth infers a layer index to each pixel, forming an interpretable image decomposition through a discrete, globally consistent ordering of elements optimized for editability. We also propose and train a neural network using a curated dataset of layered vector graphics to predict layering directly from raster inputs. Our layer index inference unlocks a range of powerful downstream applications. In particular, it significantly outperforms state-of-the-art baselines for image vectorization while also enabling high-fidelity text-to-vector-graphics generation, automatic 3D relief generation from 2D images, and intuitive depth-aware editing. By reframing depth from a physical quantity to a creative abstraction, illustrator's depth prediction offers a new foundation for editable image decomposition.

Two-dimensional ultrasound (US) imaging is widely used in abdominal interventional procedures. However, liver respiratory motion and the lack of spatial structure information in 2-D US images significantly impact the accuracy of interventions. This work aimed to construct a 3-D respiratory motion model of the liver based on optimized structural vector projection of US images to improve liver motion tracking accuracy. This work first extracts vascular topological information from 3-D US images and tracks 2-D vessel motion in 2-D US images. The 2-D vessel motions are spatially mapped to 3-D space via topological information, enabling 3-D liver motion estimation through structure vector projection optimization. To enhance estimation accuracy in sequential US images, a multi-path constrained optimization strategy is used to evaluate and select optimal 3-D motion vectors as the final estimation result. This framework ultimately establishes a 3-D respiratory motion model of the liver based on structural vector projection analysis. The proposed motion model is validated using both simulated and clinical US data. Experimental results demonstrate that the method achieves an average motion estimation error of less than 2mm and outperforms existing methods in the evaluated experiments, indicating its capability for accurate 3-D respiratory motion tracking tasks.

Vein biometrics has emerged as a promising biometric modality for personal identity authentication, benefiting from its intrinsic properties such as high discriminative capability, resistance to forgery, and contactless acquisition. Recent advances in artificial intelligence, particularly deep learning, have significantly accelerated its development. This paper presents a comprehensive and systematic survey of AI-enhanced vein biometrics. We review fundamental principles, publicly available datasets, and evaluation protocols, and systematically analyze existing methods across the entire vein biometric pipeline, including acquisition, preprocessing, feature extraction, recognition and verification, security and privacy protection, and multimodal fusion. Furthermore, we summarize representative application scenarios, identify key challenges, and highlight promising directions for future research. To facilitate reproducible research and long-term development of the field, we release an open, evolving research resource Awesome-Vein-Biometrics that systematically summarizes and tracks recent advances in vein biometrics.

As multi-object tracking (MOT) tasks continue to evolve toward more general and multi-modal scenarios, the rigid and task-specific architectures of existing MOT methods increasingly hinder their applicability across diverse tasks and limit flexibility in adapting to new tracking formulations. Most approaches rely on fixed output heads and bespoke tracking pipelines, making them difficult to extend to more complex or instruction-driven tasks. To address these limitations, we propose AR-MOT, a novel autoregressive paradigm that formulates MOT as a sequence generation task within a large language model (LLM) framework. This design enables the model to output structured results through flexible sequence construction, without requiring any task-specific heads. To enhance region-level visual perception, we introduce an Object Tokenizer based on a pretrained detector. To mitigate the misalignment between global and regional features, we propose a Region-Aware Alignment (RAA) module, and to support long-term tracking, we design a Temporal Memory Fusion (TMF) module that caches historical object tokens. AR-MOT offers strong potential for extensibility, as new modalities or instructions can be integrated by simply modifying the output sequence format without altering the model architecture. Extensive experiments on MOT17 and DanceTrack validate the feasibility of our approach, achieving performance comparable to state-of-the-art methods while laying the foundation for more general and flexible MOT systems.

This paper presents the NTIRE 2026 image super-resolution (×4) challenge, one of the associated competitions of the NTIRE 2026 Workshop at CVPR 2026. The challenge aims to reconstruct high-resolution (HR) images from low-resolution (LR) inputs generated through bicubic downsampling with a ×4 scaling factor. The objective is to develop effective super-resolution solutions and analyze recent advances in the field. To reflect the evolving objectives of image super-resolution, the challenge includes two tracks: (1) a restoration track, which emphasizes pixel-wise fidelity and ranks submissions based on PSNR; and (2) a perceptual track, which focuses on visual realism and evaluates results using a perceptual score. A total of 194 participants registered for the challenge, with 31 teams submitting valid entries. This report summarizes the challenge design, datasets, evaluation protocol, main results, and methods of participating teams. The challenge provides a unified benchmark and offers insights into current progress and future directions in image super-resolution.

Existing self-supervised methods for depth-pose joint learning mainly focus on the design of sophisticated depth estimation networks, while pose estimation is mostly conducted using a vanilla black-box CNN-based PoseNet that directly regresses the relative pose from two frames. Since direct pose regression is a non-trivial task requiring abstract understanding of the scenes, self-supervised PoseNet can hardly generalize to challenging scenarios with complex camera motions, leading to inferior performance of both pose and depth estimation. To address this issue, we propose a pose estimation method that explicitly recovers relative poses from 3D-2D correspondences, in conjunction with depth estimation via a differentiable weighted Perspective-n-Point (PnP) module. Global pose estimation from input images then essentially amounts to local image matching, which mainly relies on low-level cues and is therefore more concrete with stronger generalization ability. In addition, a weighting module is learned to adaptively filter out outliers based on their appearance feature, since not all matches contribute equally in solving relative poses, and only reliable ones are needed for robust pose recovery. We implement the above pipeline using a fully differentiable network, which allows self-supervised joint depth-pose learning in an end-to-end manner. Extensive experiments show that our method achieves stateof-the-art performance on KITTI monocular depth estimation and visual odometry benchmarks, with a 2%-5% improvement in depth accuracy (Abs Rel) using the identical model architecture and 40% improvement in pose accuracy (ATE) compared with PoseNet. Furthermore, our approach demonstrates superior generalization ability on indoor scenarios, outperforming existing self-supervised counterparts on the NYUv2 dataset by 1.5% in depth accuracy (Abs Rel).

Pansharpening aims to generate high-resolution multispectral (HRMS) images by fusing low-resolution multispectral (LRMS) with corresponding high-resolution panchromatic (PAN) images. However, the prevailing transformer-based approaches have shown limitations in capturing interaction information between spectral features, resulting in inadequate utilization of spectral information and artifacts after upsampling. The spectral angle mapper (SAM) is a spectral similarity metric that has been widely adopted in hyperspectral imaging tasks for its ability to preserve spectral fidelity. Inspired by this, we propose a novel spectral-aware attention mechanism, termed SAM self-attention (SAMSA), which introduces strong spectral inductive biases into the model, thereby facilitating more stable and effective training. Building on this, we further design a spatial-spectral reciprocal attention transformer (S-2-RAT) block that jointly computes spatial and spectral self-attentions in a reciprocal manner, enabling the two branches to complement each other's limitations. By stacking S-2-RAT blocks in a U-Net framework, we construct the S-2-reciprocal attention transformer, a dedicated spatial-spectral reciprocal attention network for pansharpening. Extensive experiments on the WorldView-3, QuickBird, and GaoFen-2 datasets demonstrate that our approach achieves superior performance, ranking first in 10 out of 12 comparisons, and reducing the SAM by 4.6% against the best competitor under reduced-resolution settings.