3887 papers
The paper proposes SemVID, a training-free token pruning framework for Video Temporal Grounding that maintains high accuracy and efficiency by allocating token budgets based on query relevance and inter-frame variation while preserving critical evidence and cross-frame connectivity through the strategic selection of object, motion, and context tokens.
This paper proposes a cardiac cine MRI reconstruction method using Gabor primitives, which combine Gaussian envelopes with complex exponentials to enable flexible k-space coverage and a low-rank spatiotemporal decomposition, achieving superior performance over compressed sensing, Gaussian primitives, and implicit neural representations while offering physically interpretable parameters.
This paper introduces OWL, a novel analytical function that enables real-time, scaled 3D scene reconstruction and camera heading estimation from raw visual motion cues alone, without requiring prior knowledge of the environment or camera motion, thereby bridging theoretical perception concepts with practical applications in robotics and autonomous navigation.
This paper introduces a novel pseudo-3D longitudinal inpainting framework based on Denoising Diffusion Probabilistic Models and Region-Aware Diffusion that significantly outperforms state-of-the-art baselines in perceptual fidelity, temporal stability, and processing speed for removing evolving lesions from brain MRI scans.
The paper introduces MultiHaystack, a new benchmark comprising over 46,000 multimodal documents, images, and videos to evaluate the critical gap between retrieval and reasoning in multimodal large language models, revealing that current systems struggle significantly when required to locate evidence within large-scale, heterogeneous corpora rather than being provided with it directly.
This paper introduces a novel framework for interpretable audiovisual geolocation that combines a high-quality global video benchmark (AVG) with a three-stage process of decomposing audio into semantic "acoustic atoms," multimodal reasoning via GRPO-finetuned MLLMs, and Riemannian Flow Matching to achieve significantly higher precision than unimodal baselines.
Any2Full is a one-stage, domain-general framework that reformulates depth completion as a scale-prompting adaptation of pretrained monocular depth estimation models via a Scale-Aware Prompt Encoder, achieving superior robustness and efficiency by eliminating the computational overhead and distortions of traditional two-stage alignment methods.
This paper proposes a lightweight, interpretable framework that extends the Discriminative Histogram of Gradient Magnitude to 3D space, combining slice-level and volume-level features with a minimal-parameter classifier to achieve robust and accurate detection of motion artifacts in structural brain MRI across diverse acquisition sites.
This paper introduces an automated, human-free pipeline using self-supervised Vision Transformers to convert the ImageNet training set into a high-quality multi-label dataset, which significantly improves both in-domain classification accuracy and downstream transfer performance compared to traditional single-label supervision.
This paper proposes a CLIP-based multi-modal framework that automatically generates structured surgical timelines and narratives by aligning video frames with textual gesture descriptions, thereby eliminating the need for time-consuming manual annotation or vague post-operative reports.
Uni-LVC is a unified learned video compression framework that integrates intra and inter coding into a single model by conditioning inter-coding on temporal cues via a cross-attention module and a reliability-aware classifier, thereby achieving superior rate-distortion performance across low-delay and random-access scenarios while maintaining computational efficiency.
This paper introduces Icarus, a deep learning-based all-weather sky model that overcomes the limitations of existing methods in handling full dynamic range and class-imbalanced solar regions to generate photorealistic, user-controllable environment maps for accurate Image-Based Lighting.
This paper introduces SCORE, a novel model merging method that resolves singular subspace conflicts between domain-specific models by projecting them into a shared orthogonal basis, thereby significantly improving generalization to unseen domains compared to existing approaches.
This paper introduces LayerBind, a training-free and plug-and-play method for Diffusion Transformers that achieves precise regional and occlusion control in text-to-image generation by modeling distinct object instances as separate layers during early denoising stages and fusing them through a semantic nursing mechanism.
The paper introduces BM25-V, a two-stage image retrieval framework that leverages sparse visual-word activations from a Sparse Auto-Encoder combined with Okapi BM25 scoring to achieve near-dense accuracy with high interpretability and computational efficiency.
This paper introduces a spectral diagnostic framework to reveal that preserving spectral structure, rather than merely enhancing spatial details, is the critical factor for achieving high-quality 3D reconstruction in 2D-to-3D pipelines, demonstrating that structural spectral consistency is the strongest predictor of novel view synthesis performance.
This paper introduces EventGeM, a state-of-the-art, real-time Visual Place Recognition system for event cameras that fuses global ViT features, local MaxViT keypoints, and depth-based structural similarity to achieve superior localization accuracy across diverse lighting conditions and benchmark datasets.
This paper introduces LIPAR, a training-free framework that accelerates video generation by pruning redundant latent patches and recovering attention values to maintain quality, thereby achieving a 1.45x throughput increase without compromising visual fidelity.
This paper introduces Margin and Consistency Supervision (MaCS), an architecture-agnostic regularization framework that combines a hinge-squared margin penalty and a consistency regularizer to simultaneously enhance the calibration, robustness, and generalization of deep vision models without requiring additional data or architectural changes.
This paper proposes a unified, single-stage architectural framework that jointly processes image and Stokes domains to achieve state-of-the-art performance in recovering polarimetric parameters from various degraded observations, including low-light noise, motion blur, and mosaicing artifacts, while ensuring physical consistency and avoiding error accumulation.