AAAI 2026 Papers

Video prediction is plagued by a fundamental trilemma: achieving high-resolution and perceptual quality typically comes at the cost of real-time speed, hindering its use in latency-critical applications. This challenge is most acute for autonomous UAVs in dense urban environments, where foreseeing events from high-resolution imagery is non-negotiable for safety. Existing methods, reliant on iterative generation (diffusion, autoregressive models) or quadratic-complexity attention, fail to meet these stringent demands on edge hardware. To break this long-standing trade-off, we introduce RAPTOR, a video prediction architecture that achieves real-time, high-resolution performance. RAPTOR’s single-pass design avoids the error accumulation and latency of iterative approaches. Its core innovation is Efficient Video Attention (EVA), a novel translator module that factorizes spatiotemporal modeling. Instead of processing flattened spacetime tokens with O((ST)^2) or O(ST) complexity, EVA alternates operations along the spatial (S) and temporal (T) axes. This factorization reduces the time complexity to O(S + T) and memory complexity to O(max(S, T)), enabling global context modeling at 512^2 resolution and beyond, operating directly on dense feature maps with a patch-free design. Complementing this architecture is a 3-stage training curriculum that progressively refines predictions from coarse structure to sharp, temporally coherent details. Experiments show RAPTOR is the first predictor to exceed 30 FPS on a Jetson AGX Orin for 512^2 video, setting a new state-of-the-art on UAVid, KTH, and a custom high-resolution dataset in PSNR, SSIM, and LPIPS. Critically, RAPTOR boosts the mission success rate in a real-world UAV navigation task by 18%, paving the way for safer and more anticipatory embodied agents.

In controllable image synthesis, generating coherent and consistent images from multiple references with spatial layout awareness remains an open challenge. We propose LAMIC, a Layout-Aware Multi-Image Composition framework that, for the first time, extends single-reference diffusion models to multi-reference scenarios in a training-free manner. Built upon the MMDiT model, LAMIC introduces two plug-and-play attention mechanisms: 1) Group Isolation Attention (GIA) to enhance entity disentanglement; and 2) Region-Modulated Attention (RMA) to enable layout-aware generation. To comprehensively evaluate model capabilities, we further introduce three metrics: 1) Inclusion Ratio (IN-R) and Fill Ratio (FI-R) for assessing layout control; and 2) Background Similarity (BG-S) for measuring background consistency. Extensive experiments show that LAMIC achieves state-of-the-art performance across most major metrics: it consistently outperforms existing multi-reference baselines in ID-S, BG-S, IN-R and AVG scores across all settings, and achieves the best DPG in complex composition tasks. These results demonstrate LAMIC's superior abilities in identity keeping, background preservation, layout control, and prompt-following, all achieved without any training or fine-tuning, showcasing strong zero-shot generalization ability. By inheriting the strengths of advanced single-reference models and enabling seamless extension to multi-image scenarios, LAMIC establishes a new training-free paradigm for controllable multi-image composition. As foundation models continue to evolve, LAMIC's performance is expected to scale accordingly.

Diffusion models have recently advanced video editing, yet controllable editing remains challenging due to the need for precise manipulation of diverse object properties. Current methods require different control signal for diverse editing tasks, which complicates model design and demands significant training resources. To address this, we propose O-DisCo-Edit, a unified framework that incorporates a novel object distortion control (O-DisCo). This signal, based on random and adaptive noise, flexibly encapsulates a wide range of editing cues within a single representation. Paired with a “copy-form” preservation module for preserving non-edited regions, O-DisCo-Edit enables efficient, high-fidelity editing through an effective training paradigm. Extensive experiments and comprehensive human evaluations consistently demonstrate that O-DisCo-Edit surpasses both specialized and multitask state-of-the-art methods across various video editing tasks.

Infrared and visible image fusion (IVIF) technology has become a frontier of great interest due to the ability to integrate information from multiple sources. However, the progressive slowdown of weight updates in deep networks (i.e., “network laziness” phenomenon), makes existing methods far from realizing the full characterization potential. To this end, we propose a lightweight fusion method for IVIF, Anti-Inert Dynamic Fusion (AIDFusion), to fully utilize the potential of the network at all levels. Specifically, by progressively regulating the collaborative Learning process of multi-level prediction in the network, Dynamic Inertia Inhibition Learning Strategy (DIILS) is proposed to adaptively and efficiently inhibit inertia accumulation. Subsequently, to deeply explore the representation potential while breaking through the performance threshold, lightweight Multi-dimensional modulation fusion module (MMFM) is specifically proposed to capture comprehensive multi-view and multi-scale features efficiently. Finally, considering the semantic bias between the prediction maps of DIILS and the fusion feature of MMFM, Fourier Analysis Convolution (FAConv) is designed in feature recovery as a bridge between prediction and fusion to accomplish the implicit periodic modeling. Based on the above study, extensive experiments on three public IVIF datasets demonstrate the dual advantages of AIDFusion in terms of fusion performance and computational overhead compared to state-of-the-art baseline methods.

Conventional document layout analysis (DLA) traditionally depends on empirical priors or a fixed set of learnable queries executed in a single forward pass. While sufficient for early-generation documents with a small, predetermined number of regions, this paradigm struggles with contemporary documents, which exhibit diverse element counts and increasingly complex layouts. To address challenges posed by modern documents, we present HybriDLA, a novel generative framework that unifies diffusion and autoregressive decoding within a single layer. The diffusion component iteratively refines bounding-box hypotheses, whereas the autoregressive component injects semantic and contextual awareness, enabling precise region prediction even in highly varied layouts. To further enhance detection quality, we design a multi-scale feature-fusion encoder that captures both fine-grained and high-level visual cues. This architecture elevates performance to 83.5% mean Average Precision (mAP). Extensive experiments on the DocLayNet and M6Doc benchmarks demonstrate that HybriDLA sets a state-of-the-art performance, outperforming previous approaches.

The rapid advancement of Artificial Intelligence Generated Content (AIGC) techniques has unlocked opportunities in generating diverse and compelling advertising images based on referenced product images and textual scene descriptions. This capability substantially reduces human labor and production costs in traditional marketing workflows. However, existing AIGC techniques either demand extensive fine-tuning for each referenced image to achieve high fidelity, or they struggle to maintain fidelity across diverse products, making them impractical for e-commerce and marketing industries. To tackle this limitation, we first construct AdProd-100K, a large-scale advertising image generation dataset. A key innovation in its construction is our dual data augmentation strategy, which fosters robust, 3D-aware representations crucial for realistic and high-fidelity image synthesis. Leveraging this dataset, we propose RefAdGen, a generation framework that achieves high fidelity through a decoupled design. The framework enforces precise spatial control by injecting a product mask at the U-Net input, and employs an efficient Attention Fusion Module (AFM) to integrate product features. This design effectively resolves the fidelity-efficiency dilemma present in existing methods. Extensive experiments demonstrate that RefAdGen achieves state-of-the-art performance, showcasing robust generalization by maintaining high fidelity and remarkable visual results for both unseen products and challenging real-world, in-the-wild images. This offers a scalable and cost-effective alternative to traditional workflows.

Training-free video object editing aims to achieve precise object-level manipulation, including object insertion, swapping, and deletion. However, it faces significant challenges in maintaining fidelity and temporal consistency. Existing methods, often designed for U-Net architectures, suffer from two primary limitations: inaccurate inversion due to first-order solvers, and contextual conflicts caused by crude "hard" feature replacement. These issues are more challenging in Diffusion Transformers (DiTs), where the unsuitability of prior layer-selection heuristics makes effective guidance challenging. To address these limitations, we introduce ContextFlow, a novel training-free framework for DiT-based video object editing. In detail, we first employ a high-order Rectified Flow solver to establish a robust editing foundation. The core of our framework is Adaptive Context Enrichment (for specifying what to edit), a mechanism that addresses contextual conflicts. Instead of replacing features, it enriches the self-attention context by concatenating Key-Value pairs from parallel reconstruction and editing paths, empowering the model to dynamically fuse information. Additionally, to determine where to apply this enrichment (for specifying where to edit), we propose a systematic, data-driven analysis to identify task-specific vital layers. Based on a novel Guidance Responsiveness Metric, our method pinpoints the most influential DiT blocks for different tasks (e.g., insertion, swapping), enabling targeted and highly effective guidance. Extensive experiments show that ContextFlow significantly outperforms existing training-free methods and even surpasses several state-of-the-art training-based approaches, delivering temporally coherent, high-fidelity results.

Video Large Language Models (Video-LLMs) have demonstrated significant potential in the areas of video captioning, search, and summarization. However, current Video-LLMs still face challenges with long real-world videos. Recent methods have introduced a retrieval mechanism that retrieves query-relevant KV caches for question answering, enhancing the efficiency and accuracy of long real-world videos. However, the compression and retrieval of KV caches are still not fully explored. In this paper, we propose StreamKV, a training-free framework that seamlessly equips Video-LLMs with advanced KV cache retrieval and compression. Compared to previous methods that used uniform partitioning, StreamKV dynamically partitions video streams into semantic segments, which better preserves semantic information. For KV cache retrieval, StreamKV calculates a summary vector for each segment to retain segment-level information essential for retrieval. For KV cache compression, StreamKV introduces a guidance prompt designed to capture the key semantic elements within each segment, ensuring only the most informative KV caches are retained for answering questions. Moreover, StreamKV unifies KV cache retrieval and compression within a single module, performing both in a layer-adaptive manner, thereby further improving the effectiveness of streaming video question answering. Extensive experiments on StreamingVQA benchmarks demonstrate that StreamKV significantly outperforms existing Online Video-LLMs, achieving superior accuracy while substantially improving both memory efficiency and computational latency.

3D Gaussian Splatting (3DGS) achieves impressive rendering fidelity and speed for novel view synthesis. However, its substantial data size poses a significant challenge for practical applications. While many compression techniques have been proposed, they fail to efficiently utilize existing bitstreams in on-demand applications due to their lack of progressivity, leading to a waste of resource. To address this issue, we propose PCGS (Progressive Compression of 3D Gaussian Splatting), which adaptively controls both the quantity and quality of Gaussians (or anchors) to enable effective progressivity for on-demand applications. For quantity, we introduce a progressive masking strategy that incrementally incorporates new anchors while refining existing ones to enhance fidelity. For quality, we propose a progressive quantization approach that gradually reduces quantization step sizes to achieve finer modeling of Gaussian attributes. Furthermore, to compact the incremental bitstreams, we leverage existing quantization results to refine probability prediction, improving entropy coding efficiency across progressive levels. PCGS achieves progressivity while maintaining compression performance comparable to SoTA non-progressive methods.

Existing approaches for the problem of ultrasound image segmentation, whether supervised or semi-supervised, are typically specialized for specific anatomical structures or tasks, limiting their practical utility in clinical settings. In this paper, we pioneer the task of universal semi-supervised ultrasound image segmentation and propose ProPL, a framework that can handle multiple organs and segmentation tasks while leveraging both labeled and unlabeled data. At its core, ProPL employs a shared vision encoder coupled with prompt-guided dual decoders, enabling flexible task adaptation through a prompting-upon-decoding mechanism and reliable self-training via an uncertainty-driven pseudo-label calibration (UPLC) module. To facilitate research in this direction, we introduce a comprehensive ultrasound dataset spanning 5 organs and 8 segmentation tasks. Extensive experiments demonstrate that ProPL outperforms state-of-the-art methods across various metrics, establishing a new benchmark for universal ultrasound image segmentation.

Variational autoencoder (VAE)-based frameworks possess a natural advantage in modeling the shared and private information inherent in multimodal data. However, current models focus on improving the quality of shared representations from the reconstruction perspective, lacking explicit mechanisms to model their underlying semantic structure. In this paper, we propose the multimodal Gaussian mixture variational autoencoder with consistency regularizations, which introduces a Gaussian mixture prior over the shared latent space to enhance its semantic structure and encourage the formation of cluster-aware latent representations. To address the cross-modal inconsistency problem under missing modality conditions, we propose a cluster-guided regularization strategy that enforces the cross-modal consistency using the pseudo-category labels from unsupervised clustering. Additionally, we design a self-supervised contrastive regularization strategy to align semantically similar representations across modalities. Extensive experiments on MNIST-SVHN and MNIST-CDCB datasets demonstrate that our method significantly outperforms prior state-of-the-art models in generation, classification, and retrieval tasks.

Open-vocabulary multi-object tracking (OV-MOT) aims to track objects with unseen categories beyond the training set. While existing methods rely on pseudo video sequences synthesized from static images, they struggle to model realistic motion patterns, resulting in limited association performance in real-world scenarios. To alleviate these issues, we propose SAM2-OV, a novel association learning-free OV-MOT method that adopts a detection-only tuning paradigm, eliminating the need for synthetic sequences or spatiotemporal supervision and substantially reducing the overall learnable parameters. The core of our method is a Unified Detection Module (UDM), which effectively provides object-level prompts to enable SAM2 for OV-MOT. Enabled by UDM, SAM2-OV is the first to integrate SAM2 for OV-MOT, fully unleashing its zero-shot cross-frame association ability. To further enhance object association under occlusion and abrupt motion, we introduce a Motion Prior Assistance Module (MPAM) that incorporates motion cues into the mask selection process. In addition, a Semantic Enhancement Adapter (SEA) distilled from CLIP is used to improve classification generalization. A sparse prompting strategy is also adopted to reduce computational redundancy by triggering detection only on selected keyframes. As only the detection module is tuned on static images, the overall training process remains simple and efficient. Experiments on the TAO dataset demonstrate that SAM2-OV achieves state-of-the-art performance under the TETA metric, particularly on novel categories. Evaluations on the KITTI dataset show the strong zero-shot cross-domain transferability of our SAM2-OV.

Normalizing Flows (NFs) are a class of generative models distinguished by a mathematically invertible architecture, where the forward pass transforms data into a latent space for density estimation, and the reverse pass generates new samples from this space. This characteristic creates an intrinsic synergy between representation learning and data generation. However, the generative quality of standard NFs is limited by poor semantic representations from log-likelihood optimization. To remedy this, we propose a novel alignment strategy that creatively leverages the invertibility of NFs: instead of regularizing the forward pass, we align the intermediate features of the generative (reverse) pass with representations from a powerful vision foundation model, demonstrating superior effectiveness over naive alignment. We also introduce a novel training-free, test-time optimization algorithm for classification, which provides a more intrinsic evaluation of the NF's embedded semantic knowledge. Comprehensive experiments demonstrate that our approach accelerates the training of NFs by over 3.3x, while simultaneously delivering significant improvements in both generative quality and classification accuracy. New state-of-the-art results for NFs are established on ImageNet 64 x 64 and 256 x 256.

Unsupervised visual anomaly detection from multi-view images presents a significant challenge: distinguishing genuine defects from benign appearance variations caused by viewpoint changes. Existing methods, often designed for single-view inputs, treat multiple views as a disconnected set of images, leading to inconsistent feature representations and a high false-positive rate. To address this, we introduce ViewSense-AD (VSAD), a novel framework that learns viewpoint-invariant representations by explicitly modeling geometric consistency across views. At its core is our Multi-View Alignment Module (MVAM), which leverages homography to project and align corresponding feature regions between neighboring views. We integrate MVAM into a View-Align Latent Diffusion Model (VALDM), enabling progressive and multi-stage alignment during the denoising process. This allows the model to build a coherent and holistic understanding of the object's surface from coarse to fine scales. Furthermore, a lightweight Fusion Refiner Module (FRM) enhances the global consistency of the aligned features, suppressing noise and improving discriminative power. Anomaly detection is performed by comparing multi-level features from the diffusion model against a learned memory bank of normal prototypes. Extensive experiments on the challenging RealIAD and MANTA datasets demonstrate that VSAD sets a new state-of-the-art, significantly outperforming existing methods in pixel, view, and sample-level visual anomaly detection, proving its robustness to large viewpoint shifts and complex textures.

Visible-Infrared (RGB-IR) Unmanned Aerial Vehicle (UAV) object detection integrates complementary cues from visible and infrared sensors, offering broad application potential. However, due to sensor parallax, it still faces the challenge of weak spatial misalignment, which significantly limits its performance in UAV-based object detection. Existing methods emphasize strict alignment, overlooking spectral heterogeneity under varying illumination. To address these issues, we propose the Illumination Guided Implicit Alignment Network (IGIANet) to mitigate modality heterogeneity without explicit alignment. Specifically, we integrate three novel modules. First, we propose an illumination-guided frequency modulation module that adaptively allocates fusion weights to visible and infrared features based on global illumination estimation, effectively alleviating modality imbalance under varying lighting conditions. Second, we introduce a frequency-guided cross-modality differential enhancement module, which computes differential cues across frequency domains to enhance complementary information and highlight weakly aligned and low-contrast regions. Finally, we introduce an implicit alignment-driven dynamic fusion module that actively estimates offsets and generates dynamic, position-adaptive fusion kernels to align and fuse modalities. Extensive experiments demonstrate that IGIANet outperforms state-of-the-art models on various benchmarks, achieving 80.9% mAP on DroneVehicle, 57.1% mAP on VEDAI, and 49.4% mAP on FLIR.

Endoscopic images often suffer from diverse and co-occurring degradations such as low lighting, smoke, and bleeding, which obscure critical clinical details. Existing restoration methods are typically task-specific and often require prior knowledge of the degradation type, limiting their robustness in real-world clinical use. We propose EndoIR, an all-in-one, degradation-agnostic diffusion-based framework that restores multiple degradation types using a single model. EndoIR introduces a Dual-Domain Prompter that extracts joint spatial–frequency features, coupled with an adaptive embedding that encodes both shared and task-specific cues as conditioning for denoising. To mitigate feature confusion in conventional concatenation-based conditioning, we design a Dual-Stream Diffusion architecture that processes clean and degraded inputs separately, with a Rectified Fusion Block integrating them in a structured, degradation-aware manner. Furthermore, Noise-Aware Routing Block improves efficiency by dynamically selecting only noise-relevant features during denoising. Experiments on SegSTRONG-C and CEC datasets demonstrate that EndoIR achieves state-of-the-art performance across multiple degradation scenarios while using fewer parameters than strong baselines, and downstream segmentation experiments confirm its clinical utility.

While hyperspectral images (HSI) benefit from numerous spectral channels that provide rich information for classification, the increased dimensionality and sensor variability make them more sensitive to distributional discrepancies across domains, which in turn can affect classification performance. To tackle this issue, hyperspectral single-source domain generalization (SDG) typically employs data augmentation to simulate potential domain shifts and enhance model robustness under the condition of single-source domain training data availability. However, blind augmentation may produce samples misaligned with real-world scenarios, while excessive emphasis on realism can suppress diversity, highlighting a tradeoff between realism and diversity that limits generalization to target domains. To address this challenge, we propose a spectral property-driven data augmentation (SPDDA) that explicitly accounts for the inherent properties of HSI, namely the device-dependent variation in the number of spectral channels and the mixing of adjacent channels. Specifically, SPDDA employs a spectral diversity module that resamples data from the source domain along the spectral dimension to generate samples with varying spectral channels, and constructs a channel-wise adaptive spectral mixer by modeling inter-channel similarity, thereby avoiding fixed augmentation patterns. To further enhance the realism of the augmented samples, we propose a spatial-spectral co-optimization mechanism, which jointly optimizes a spatial fidelity constraint and a spectral continuity self-constraint. Moreover, the weight of the spectral self-constraint is adaptively adjusted based on the spatial counterpart, thus preventing over-smoothing in the spectral dimension and preserving spatial structure. Extensive experiments conducted on three remote sensing benchmarks demonstrate that SPDDA outperforms state-of-the-art methods.

Biologically plausible and energy-efficient frameworks such as Spiking Neural Networks (SNNs) have not been sufficiently explored in low-level vision tasks. Taking image deraining as an example, this study addresses the representation of the inherent high-pass characteristics of spiking neurons, specifically in image deraining and innovatively proposes the Visual LIF (VLIF) neuron, overcoming the obstacle of lacking spatial contextual understanding present in traditional spiking neurons. To tackle the limitation of frequency-domain saturation inherent in conventional spiking neurons, we leverage the proposed VLIF to introduce the Spiking Decomposition and Enhancement Module and the lightweight Spiking Multi-scale Unit for hierarchical multi-scale representation learning. Extensive experiments across five benchmark deraining datasets demonstrate that our approach significantly outperforms state-of-the-art SNN-based deraining methods, achieving this superior performance with only 13% of their energy consumption. These findings establish a solid foundation for deploying SNNs in high-performance, energy-efficient low-level vision tasks.

Camera-based 3D semantic scene completion (SSC) provides dense geometric and semantic perception for autonomous driving and robotic navigation. However, existing methods rely on a coupled encoder to deliver both semantic and geometric priors, which forces the model to make a trade-off between conflicting demands and limits its overall performance. To tackle these challenges, we propose FoundationSSC, a novel framework that performs dual decoupling at both the source and pathway levels. At the source level, we introduce a foundation encoder that provides rich semantic feature priors for the semantic branch and high-fidelity stereo cost volumes for the geometric branch. At the pathway level, these priors are refined through specialised, decoupled pathways, yielding superior semantic context and depth distributions. Our dual-decoupling design produces disentangled and refined inputs, which are then utilised by a hybrid view transformation to generate complementary 3D features. Additionally, we introduce a novel Axis-Aware Fusion (AAF) module that addresses the often-overlooked challenge of fusing these features by anisotropically merging them into a unified representation. Extensive experiments demonstrate the advantages of FoundationSSC, achieving simultaneous improvements in both semantic and geometric metrics, surpassing prior bests by +0.23 mIoU and +2.03 IoU on SemanticKITTI. Additionally, we achieve state-of-the-art performance on SSCBench-KITTI-360, with 21.78 mIoU and 48.61 IoU.