论文检索

Zhirui Kuai, Huan Zhang, Yang Yang, Yiping Ma, Mingjing Huang, Ning Gui, Li Kuang

Multimodal data significantly improves the performance of pretrained models, but its practical application is often limited by missing or incomplete data across modalities. There are two key challenges that existing methods of synthesizing missing data face: (1) semantic inaccuracies due to model hallucinations and (2) discrepancies in distribution preferences between generated and original data. To address these challenges, we propose a novel three-stage multimodal data augmentation framework (GFR), which Generate, Filter, and Rank missing modality data. Our framework leverages multimodal large models for diverse data generation, designs a scene graph matching-based filtering algorithm to ensure semantic consistency, and constructs a preference-aware ranking model to align the generated data with both the original distribution and task relevance. Our framework not only enhances semantic diversity and consistency in data generation but also effectively captures the implicit characteristics of the original dataset and the target model. We demonstrate the effectiveness of GFR across multiple datasets by testing different missing types and missing ratios.

Junseo Koo, Jinseo Jeong, Gunhee Kim

The recent introduction of 3D Gaussian Splatting (3DGS) has significantly advanced novel view synthesis. Several studies have further improved the rendering quality of 3DGS, yet they still exhibit noticeable visual discrepancies when synthesizing views at sampling rates unseen during training. Specifically, they suffer from (i) erosion-induced blurring artifacts when zooming in and (ii) dilation-induced staircase artifacts when zooming out. We speculate that these artifacts arise from the fundamental limitation of the alpha blending adopted in 3DGS methods. Instead of the conventional alpha blending that computes alpha and transmittance as scalar quantities over a pixel, we propose to replace it with our novel Gaussian Blending that treats alpha and transmittance as spatially varying distributions. Thus, transmittances can be updated considering the spatial distribution of alpha values across the pixel area, allowing nearby background splats to contribute to the final rendering. Our Gaussian Blending maintains real-time rendering speed and requires no additional memory cost, while being easily integrated as a drop-in replacement into existing 3DGS-based or other NVS frameworks. Extensive experiments demonstrate that Gaussian Blending effectively captures fine details at various sampling rates unseen during training, consistently outperforming existing novel view synthesis models across both unseen and seen sampling rates.

Xiaohui Kong, Qian Liu, Dandan Zhu, Kaiwei Zhang, Xiongkuo Min

Scanpath prediction in omnidirectional images (ODIs) serves as a critical component for optimizing foveated rendering efficiency and enhancing interactive quality in virtual reality systems. However, existing scanpath prediction methods for ODIs still suffer from fundamental limitations: (1) inadequate modeling and capturing of long-range temporal dependencies in fixation regions, and (2) suboptimal integration of spatial and temporal visual features, ultimately compromising prediction performance. To address these limitations, we propose a novel Dual-Temporal Modulated Diffusion model for Omnidirectional Images Scanpath Prediction, named SalDiff-DTM model, to effectively generate realistic human eye viewing trajectories. Specifically, to effectively model spatial relationships, we propose a novel Dual-Graph Convolutional Network (Dual-GCN) module that simultaneously captures semantic-level and image-level correlations. By integrating both local spatial details and global contextual information across the internal temporal dimension, this module achieves comprehensive and robust modeling of spatial relationships. To further enhance the modeling of temporal dependencies inherent in diverse fixation patterns, we introduce TABiMamba (Temporal-Aware BiLSTM-Mamba), a dedicated module that synergistically combines the contextual sensitivity of BiLSTM with the long-range sequence modeling capabilities of Mamba. This design facilitates deep information flow and context-aware sequential reasoning, thereby enabling high-fidelity capture of intricate temporal correlations. Inspired by the progressive refinement mechanism of diffusion models in various generative tasks, we propose a saliency-guided diffusion module that formulates the prediction problem as a conditional generative process, iteratively yielding accurate and perceptually plausible scanpaths. Extensive experiments demonstrate that SalDiff-DTM significantly outperforms state-of-the-art models, paving the way for future advancements in eye-tracking technologies and cognitive modeling.

Dequan Kong, Honghua Chen, Zhe Zhu, Mingqiang Wei

Existing diffusion-based 3D shape completion methods typically use a conditional paradigm, injecting incomplete shape information into the denoising network via deep feature interactions (e.g., concatenation, cross-attention) to guide sampling toward complete shapes, often represented by voxel-based distance functions. However, these approaches fail to explicitly model the optimal global transport path, leading to suboptimal completions. Moreover, performing diffusion directly in voxel space imposes resolution constraints, limiting the generation of fine-grained geometric details. To address these challenges, we propose BridgeShape, a novel framework for 3D shape completion via latent diffusion Schrödinger bridge. The key innovations lie in two aspects: (i) BridgeShape formulates shape completion as an optimal transport problem, explicitly modeling the transition between incomplete and complete shapes to ensure a globally coherent transformation. (ii) We introduce a Depth-Enhanced Vector Quantized Variational Autoencoder (VQ-VAE) to encode 3D shapes into a compact latent space, leveraging self-projected multi-view depth information enriched with strong DINOv2 features to enhance geometric structural perception. By operating in a compact yet structurally informative latent space, BridgeShape effectively mitigates resolution constraints and enables more efficient and high-fidelity 3D shape completion. BridgeShape achieves state-of-the-art performance on 3D shape completion benchmarks, demonstrating superior fidelity at higher resolutions and for unseen object classes.

Taeyeong Kim, SeungJoon Lee, Jung Uk Kim, MyeongAh Cho

Domain generalization in semantic segmentation faces challenges from domain shifts, particularly under adverse conditions. While diffusion-based data generation methods show promise, they introduce inherent misalignment between generated images and semantic masks. This paper presents FLEX-Seg (FLexible Edge eXploitation for Segmentation), a framework that transforms this limitation into an opportunity for robust learning. FLEX-Seg comprises three key components: (1) Granular Adaptive Prototypes that captures boundary characteristics across multiple scales, (2) Uncertainty Boundary Emphasis that dynamically adjusts learning emphasis based on prediction entropy, and (3) Hardness-Aware Sampling that progressively focuses on challenging examples. By leveraging inherent misalignment rather than enforcing strict alignment, FLEX-Seg learns robust representations while capturing rich stylistic variations. Experiments across five real-world datasets demonstrate consistent improvements over state-of-the-art methods, achieving 2.44% and 2.63% mIoU gains on ACDC and Dark Zurich. Our findings validate that adaptive strategies for handling imperfect synthetic data lead to superior domain generalization.

Seungjae Kim, SeungJoon Lee, MyeongAh Cho

Multi-object tracking (MOT) predominantly follows the tracking-by-detection paradigm, where Kalman filters serve as the standard motion predictor due to computational efficiency but inherently fail on non-linear motion patterns. Conversely, recent data-driven motion predictors capture complex non-linear dynamics but suffer from limited domain generalization and computational overhead. Through extensive analysis, we reveal that even in datasets dominated by non-linear motion, Kalman filter outperforms data-driven predictors in up to 34% of cases, demonstrating that real-world tracking scenarios inherently involve both linear and non-linear patterns. To leverage this complementarity, we propose PlugTrack, a novel framework that adaptively fuses Kalman filter and data-driven motion predictors through multi-perceptive motion understanding. Our approach employs multi-perceptive motion analysis to generate adaptive blending factors. PlugTrack achieves significant performance gains on MOT17/MOT20 and state-of-the-art on DanceTrack without modifying existing motion predictors. To the best of our knowledge, PlugTrack is the first framework to bridge classical and modern motion prediction paradigms through adaptive fusion in MOT.

Namyup Kim, Jinsung Lee, Suha Kwak

Generalized referring image segmentation (RIS) aims to segment regions in an image described by a natural language expression, handling not only single-target but also no- and multi-target scenarios. Previous approaches have proposed new components that enable a conventional RIS model to handle these additional scenarios, such as a target presence prediction head for no-target scenarios and multiple mask candidates for multi-target cases. However, we observe that these methods predominantly rely on the conventional RIS backbone without fully integrating the additional components and thus still struggle in such general scenarios. To address this, we propose an effective framework specifically tailored to handle no-target and multi-target scenarios, incorporating both architectural and data-driven approaches. Our architecture employs a learnable query designed to understand both target presence and plurality. While this approach alone outperforms previous state-of-the-art methods with similar computational requirements, we further introduce a novel data augmentation strategy that enables our framework to surpass computationally intensive LMM-based approaches.

Minjun Kim, Jaeri Lee, Jongjin Kim, Jeongin Yun, Yongmo Kwon, U Kang

How can we accurately quantize a pre-trained Vision Transformer model? Quantization algorithms compress Vision Transformers (ViTs) into low-bit formats, reducing memory and computation demands with minimal accuracy degradation. However, existing methods rely on uniform precision, ignoring the diverse sensitivity of ViT components to quantization. Metric-based Mixed Precision Quantization (MPQ) is a promising alternative, but previous MPQ methods for ViTs suffer from three major limitations: 1) coarse granularity, 2) mismatch in metric scale across component types, and 3) quantization-unaware bit allocation. In this paper, we propose LampQ (Layer-wise Mixed Precision Quantization for Vision Transformers), an accurate metric-based MPQ method for ViTs to overcome these limitations. LampQ performs layer-wise quantization to achieve both fine-grained control and efficient acceleration, incorporating a type-aware Fisher-based metric to measure sensitivity. Then, LampQ assigns bit-widths optimally through integer linear programming and further updates them iteratively. Extensive experiments show that LampQ provides the state-of-the-art performance in quantizing ViTs pre-trained on various tasks such as image classification, object detection, and zero-shot quantization.

Kwanyoung Kim

Diffusion models have demonstrated strong generative performance when using guidance methods such as classifier-free guidance (CFG), which enhance output quality by modifying the sampling trajectory. These methods typically improve a target output by intentionally degrading another, often the unconditional output, using heuristic perturbation functions such as identity mixing or blurred conditions. However, these approaches lack a principled foundation and rely on manually designed distortions. In this work, we propose Adversarial Sinkhorn Attention Guidance (ASAG), a novel method that reinterprets attention scores in diffusion models through the lens of optimal transport and intentionally increases the transport cost to disrupt unreliable attention flows. Instead of naively corrupting the attention mechanism, ASAG injects an adversarial cost within self-attention layers to reduce pixel-wise similarity between queries and keys. This deliberate degradation weakens misleading attention alignments and leads to improved conditional and unconditional sample quality. ASAG shows consistent improvements in text-to-image diffusion, and enhances controllability and fidelity in downstream applications such as IP-Adapter and ControlNet. The method is lightweight, plug-and-play, and improves reliability without requiring any model retraining.

Jaeyoon Kim, Yoonki Cho, Sung-Eui Yoon

Visual Place Recognition (VPR) has advanced significantly with high-capacity foundation models like DINOv2, achieving remarkable performance. Nonetheless, their substantial computational cost makes deployment on resource-constrained devices impractical. In this paper, we introduce an efficient asymmetric VPR framework that incorporates a high-capacity gallery model for offline feature extraction with a lightweight query network for online processing. A key challenge in this setting is ensuring compatibility between these heterogeneous networks, which conventional approaches address through computationally expensive k-NN-based compatible training. To overcome this, we propose a geographical memory bank that structures gallery features using geolocation metadata inherent in VPR databases, eliminating the need for exhaustive k-NN computations. Additionally, we introduce an implicit embedding augmentation technique that enhances the query network to model feature variations despite its limited capacity. Extensive experiments demonstrate that our method not only significantly reduces computational costs but also outperforms existing asymmetric retrieval techniques, establishing a new aspect for VPR in resource-limited environments.

Hyeonjae Kim, Dongjin Kim, Eugene Jin, Tae Hyun Kim

While deep learning-based super-resolution (SR) methods have shown impressive outcomes with synthetic degradation scenarios such as bicubic downsampling, they frequently struggle to perform well on real-world images that feature complex, nonlinear degradations like noise, blur, and compression artifacts. Recent efforts to address this issue have involved the painstaking compilation of real low-resolution (LR) and high-resolution (HR) image pairs, usually limited to several specific downscaling factors. To address these challenges, our work introduces a novel framework capable of synthesizing authentic LR images from a single HR image by leveraging the latent degradation space with flow matching. Our approach generates LR images with realistic artifacts at unseen degradation levels, which facilitates the creation of large-scale, real-world SR training datasets. Comprehensive quantitative and qualitative assessments verify that our synthetic LR images accurately replicate real-world degradations. Furthermore, both traditional and arbitrary-scale SR models trained using our datasets consistently yield much better HR outcomes.

Geewook Kim, Minjoon Seo

We propose an efficient framework to compress massive video-frame features before feeding them into large multimodal models, thereby mitigating the severe token explosion arising from hour-long videos. Our design leverages a bidirectional state-space model equipped with a gated skip connection and a learnable weighted-average pooling mechanism applied to periodically inserted learned queries. This structure enables hierarchical downsampling across both spatial and temporal dimensions, preserving performance in a cost-effective manner. Across challenging hour-long video understanding tasks, our approach demonstrates competitive results against state-of-the-art models, while significantly reducing overall token budget. Notably, replacing our state-space model with conventional modules results in substantial performance degradation, highlighting the advantages of the proposed state-space modeling for effectively compressing multi-frame video information. Our framework emphasizes resource-conscious efficiency, making it practical for real-world deployments. We validate its scalability and generality across multiple benchmarks, achieving the dual objectives of efficient resource usage and comprehensive video understanding.

Da-Yeong Kim, Yeong-Jun Cho

Point cloud completion aims to recover missing geometric structures from incomplete 3D scans, which often suffer from occlusions or limited sensor viewpoints. Existing methods typically assume fixed input/output densities or rely on image-based representations, making them less suitable for real-world scenarios with variable sparsity and limited supervision. In this paper, we introduce Density-agnostic and Class-aware Network (DANCE), a novel framework that completes only the missing regions while preserving the observed geometry. DANCE generates candidate points via ray-based sampling from multiple viewpoints. A transformer decoder then refines their positions and predicts opacity scores, which determine the validity of each point for inclusion in the final surface. To incorporate semantic guidance, a lightweight classification head is trained directly on geometric features, enabling category-consistent completion without external image supervision. Extensive experiments on the PCN and MVP benchmarks show that DANCE outperforms state-of-the-art methods in accuracy and structural consistency, while remaining robust to varying input densities and noise levels.

Changjin Kim, HyeokJun Lee, YoungJoon Yoo

Recent image denoising methods have leveraged generative modeling for real noise synthesis to address the costly acquisition of real-world noisy data. However, these generative models typically require camera metadata and extensive target-specific noisy-clean image pairs, often showing limited generalization between settings. In this paper, to mitigate the prerequisites, we propose a Single-Pair Guided Diffusion for generalized noise synthesis(GuidNoise), which uses a single noisy/clean pair as the guidance, often easily obtained by itself within a training set. To train GuidNoise, which generates synthetic noisy images from the guidance, we introduce a guidance-aware affine feature modification (GAFM) and a noise-aware refine loss to leverage the inherent potential of diffusion models. This loss function refines the diffusion model’s backward process, making the model more adept at generating realistic noise distributions. The GuidNoise synthesizes high-quality noisy images under diverse noise environments without additional metadata during both training and inference. Additionally, GuidNoise enables the efficient generation of noisy-clean image pairs at inference time, making synthetic noise readily applicable for augmenting train- ing data. This self-augmentation significantly improves denoising performance, especially in practical scenarios with lightweight models and limited training data.

Zhihui Ke, Yvyang Liu, Xiaobo Zhou, Tie Qiu

Streaming free-viewpoint video (FVV) in real-time still faces significant challenges, particularly in training, rendering, and transmission efficiency. Harnessing superior performance of 3D Gaussian Splatting (3DGS), recent 3DGS-based FVV methods have achieved notable breakthroughs in both training and rendering. However, the storage requirements of these methods can reach up to 10MB per frame, making stream FVV in real-time impossible. To address this problem, we propose a novel FVV representation, dubbed StreamSTGS, designed for real-time streaming. StreamSTGS represents a dynamic scene using canonical 3D Gaussians, temporal features, and a deformation field. For high compression efficiency, we encode canonical Gaussian attributes as 2D images and temporal features as a video. This design not only enables real-time streaming, but also inherently supports adaptive bitrate control based on network condition without any extra training. Moreover, we propose a sliding window scheme to aggregate adjacent temporal features to learn local motions, and then introduce a transformer-guided auxiliary training module to learn global motions. On diverse FVV benchmarks, StreamSTGS demonstrates competitive performance on all metrics compared to state-of-the-art methods. Notably, StreamSTGS increases the PSNR by an average of 1dB while reducing the average frame size to just 170KB.

Arjun Ramesh Kaushik, Naresh Kumar Devulapally, Vishnu Suresh Lokhande, Nalini K. Ratha, Venu Govindaraju

Custom Diffusion Models (CDMs) offer impressive capabilities for personalization in generative modeling, yet they remain vulnerable to catastrophic forgetting when learning new concepts sequentially. Existing approaches primarily focus on minimizing interference between concepts, often neglecting the potential for positive inter-concept interactions. In this work, we present Forget Less by Learning from Parents (FLLP), a novel framework that introduces a parent-child inter-concept learning mechanism in hyperbolic space to mitigate forgetting. By embedding concept representations within a Lorentzian manifold, naturally suited to modeling tree-like hierarchies, we define parent-child relationships in which previously learned concepts serve as guidance for adapting to new ones. Our method not only preserves prior knowledge but also supports continual integration of new concepts. We validate FLLP on three public datasets and one synthetic benchmark, showing consistent improvements in both robustness and generalization.

Junyong Kang, Seohyun Lim, Kyungjune Baek, Hyunjung Shim

Aligning text-to-image (T2I) diffusion models with human preferences has emerged as a critical research challenge. While Direct Preference Optimization (DPO) has established a foundation for preference learning in large language models (LLMs), its extension to diffusion models remains limited in alignment performance. In this work, we propose an enhanced version of Diffusion-DPO by introducing a stable reference model update strategy. This strategy facilitates the exploration of better alignment solutions while maintaining training stability. Moreover, we design a timestep-aware optimization strategy that further boosts performance by addressing preference learning imbalance across timesteps. Through the synergistic combination of our exploration and timestep-aware optimization, our method significantly improves the alignment performance of Diffusion-DPO on human preference evaluation benchmarks, achieving state-of-the-art results.

Donghwa Kang, Doohyun Kim, Sang-Ki Ko, Jinkyu Lee, Hyeongboo Baek, Brent ByungHoon Kang

State-of-the-art (SOTA) gradient-based adversarial attacks on spiking neural networks (SNNs), which largely rely on extending FGSM and PGD frameworks, face a critical limitation: substantial attack latency from multi-timestep processing, rendering them infeasible for practical real-time applications. This inefficiency stems from their design as direct extensions of ANN paradigms, which fail to exploit key SNN properties. In this paper, we propose the timestep compressed attack (TCA), a novel framework that significantly reduces attack latency. TCA introduces two components founded on key insights into SNN behavior. First, timestep-level backpropagation (TLBP) is based on our finding that global temporal information in backpropagation to generate perturbations is not critical for an attack’s success, enabling per-timestep evaluation for early stopping. Second, adversarial membrane potential reuse (A-MPR) is motivated by the observation that initial timesteps are inefficiently spent accumulating membrane potential, a warm-up phase that can be pre-calculated and reused. Our experiments on VGG-11 and ResNet-17 with the CIFAR-10/100 and CIFAR10-DVS datasets show that TCA significantly reduces the required attack latency by up to 56.6% and 57.1% compared to SOTA methods in white-box and black-box settings, respectively, while maintaining a comparable attack success rate.

Dachun Kai, Zeyu Xiao, Huyue Zhu, Jiaxiao Wang, Yueyi Zhang, Xiaoyan Sun

This paper addresses low-light video super-resolution (LVSR), aiming to restore high-resolution videos from low-light, low-resolution (LR) inputs. Existing LVSR methods often struggle to recover fine details due to limited contrast and insufficient high-frequency information. To overcome these challenges, we present RetinexEVSR, the first event-driven LVSR framework that leverages high-contrast event signals and Retinex-inspired priors to enhance video quality under low-light scenarios. Unlike previous approaches that directly fuse degraded signals, RetinexEVSR introduces a novel bidirectional cross-modal fusion strategy to extract and integrate meaningful cues from noisy event data and degraded RGB frames. Specifically, an illumination-guided event enhancement module is designed to progressively refine event features using illumination maps derived from the Retinex model, thereby suppressing low-light artifacts while preserving high-contrast details. Furthermore, we propose an event-guided reflectance enhancement module that utilizes the enhanced event features to dynamically recover reflectance details via a multi-scale fusion mechanism. Experimental results show that our RetinexEVSR achieves state-of-the-art performance on three datasets. Notably, on the SDSD benchmark, our method can get up to 2.95 dB gain while reducing runtime by 65% compared to prior event-based methods.

Sun Jo, Seok Young Hong, Jinhyun Kim, Seungmin Kang, Ahjin Choi, Don-Gwan An, Simon Song, Je Hyeong Hong

4D flow magnetic resonance imaging (MRI) is a reliable, non-invasive approach for estimating blood flow velocities, vital for cardiovascular diagnostics. Unlike conventional MRI focused on anatomical structures, 4D flow MRI requires high spatiotemporal resolution for early detection of critical conditions such as stenosis or aneurysms. However, achieving such resolution typically results in prolonged scan times, creating a trade-off between acquisition speed and prediction accuracy. Recent studies have leveraged physics-informed neural networks (PINNs) for super-resolution of MRI data, but their practical applicability is limited as the prohibitively slow training process must be performed for each patient. To overcome this limitation, we propose PINGS-X, a novel framework modeling high-resolution flow velocities using axes-aligned spatiotemporal Gaussian representations. Inspired by the effectiveness of 3D Gaussian splatting (3DGS) in novel view synthesis, PINGS-X extends this concept through several non-trivial novel innovations: (i) normalized Gaussian splatting with a formal convergence guarantee, (ii) axes-aligned Gaussians that simplify training for high-dimensional data while preserving accuracy and the convergence guarantee, and (iii) a Gaussian merging procedure to prevent degenerate solutions and boost computational efficiency. Experimental results on computational fluid dynamics (CFD) and real 4D flow MRI datasets demonstrate that PINGS-X substantially reduces training time while achieving superior super-resolution accuracy.