wovepaper — the arXiv observatory

MAVIN is a framework for generating multi-shot audio‑visual content with fine‑grained narrative control, using boundary‑aware attention to align temporal segments and ID‑aware prop…

Publications (100)

MotionVLA: Vision-Language-Action Model for Humanoid Motion

Reconstructing Satellites in 3D from Amateur Telescope Images

SymmeTac: Symmetric Color LED Driven Efficient Photometric Stereo Reconstruction Methods for Camera-based Tactile Sensors

Language-guided Image Reflection Separation

GeneralVLA-2: Geometry-Aware Reconstruction and Governed Memory for Robot Planning

High-Resolution Single-Shot Polarimetric Imaging Made Easy

NeIF: Representing General Reflectance as Neural Intrinsics Fields for Uncalibrated Photometric Stereo

Robust Student Network Learning

Learning to Deblur Polarized Images

Stokes-Informed Diffusion for Robust Linear Polarization Estimation

DreamPolish: Domain Score Distillation With Progressive Geometry Generation

MAVIN: Multi-Shot Audio-Visual Generation with Customized Narrative Control

Dark-EvGS: Event Camera as an Eye for Radiance Field in the Dark

AdaptiveAE: An Adaptive Exposure Strategy for HDR Capturing in Dynamic Scenes

EvPlug: Learn a Plug-and-Play Module for Event and Image Fusion

PolGS++: Physically-Guided Polarimetric Gaussian Splatting for Fast Reflective Surface Reconstruction

PIDSR: Complementary Polarized Image Demosaicing and Super-Resolution

AdderNet: Do We Really Need Multiplications in Deep Learning?

Event-driven Video Frame Synthesis

Affective Image Editing: Shaping Emotional Factors via Text Descriptions

Distilling portable Generative Adversarial Networks for Image Translation

PolarAnything: Diffusion-based Polarimetric Image Synthesis

Complementing Event Streams and RGB Frames for Hand Mesh Reconstruction

DIST: Rendering Deep Implicit Signed Distance Function with Differentiable Sphere Tracing

NeuralMPS: Non-Lambertian Multispectral Photometric Stereo via Spectral Reflectance Decomposition

V2V: Scaling Event-Based Vision through Efficient Video-to-Voxel Simulation

Deep Learning Methods for Calibrated Photometric Stereo and Beyond

NeRSP: Neural 3D Reconstruction for Reflective Objects with Sparse Polarized Images

PanoWan: Lifting Diffusion Video Generation Models to 360Â° with Latitude/Longitude-aware Mechanisms

Real-time 3D-aware Portrait Video Relighting

EvalMVX: A Unified Benchmarking for Neural 3D Reconstruction under Diverse Multiview Setups

Reconstruction as a Bridge for Event-Based Visual Question Answering

A Benchmark and Multi-Agent System for Instruction-driven Cinematic Video Compilation

Lift3D-VLA: Lifting VLA Models to 3D Geometry and Dynamics-Aware Manipulation

Towards Deeper Emotional Reflection: Crafting Affective Image Filters with Generative Priors

The NeRF Signature: Codebook-Aided Watermarking for Neural Radiance Fields

EvHandPose: Event-based 3D Hand Pose Estimation with Sparse Supervision

Self-Supervised Scene Dynamic Recovery from Rolling Shutter Images and Events

L-CAD: Language-based Colorization with Any-level Descriptions using Diffusion Priors

CRRN: Multi-Scale Guided Concurrent Reflection Removal Network

Edge-preserving Near-light Photometric Stereo with Neural Surfaces

Saliency Detection by Forward and Backward Cues in Deep-CNNs

On Positive-Unlabeled Classification in GAN

Surface Geometry Processing: An Efficient Normal-based Detail Representation

Video Generation Models Are Inherent Lighting Estimators

Hyperspectral City V1.0 Dataset and Benchmark

Personalized Image Filter: Mastering Your Photographic Style

BokehDiff: Neural Lens Blur with One-Step Diffusion

Self-calibrating Deep Photometric Stereo Networks

ReContraster: Making Your Posters Stand Out with Regional Contrast

SPLINE-Net: Sparse Photometric Stereo through Lighting Interpolation and Normal Estimation Networks

Data-Free Learning of Student Networks

Asynchronous Event Error-Minimizing Noise for Safeguarding Event Dataset

PolarVLM: Bridging the Semantic-Physical Gap in Vision-Language Models

Label Distribution Amendment with Emotional Semantic Correlations for Facial Expression Recognition

STAGE: Storyboard-Anchored Generation for Cinematic Multi-shot Narrative

EvTurb: Event Camera Guided Turbulence Removal

SpikeNVS: Enhancing Novel View Synthesis from Blurry Images via Spike Camera

EventAid: Benchmarking Event-aided Image/Video Enhancement Algorithms with Real-captured Hybrid Dataset

DANI-Net: Uncalibrated Photometric Stereo by Differentiable Shadow Handling, Anisotropic Reflectance Modeling, and Neural Inverse Rendering

Deep Photometric Stereo for Non-Lambertian Surfaces

DeRenderNet: Intrinsic Image Decomposition of Urban Scenes with Shape-(In)dependent Shading Rendering

Face Image Reflection Removal

PolGS: Polarimetric Gaussian Splatting for Fast Reflective Surface Reconstruction

Spin-UP: Spin Light for Natural Light Uncalibrated Photometric Stereo

Spatially-Varying Outdoor Lighting Estimation from Intrinsics

Colorizing Monochromatic Radiance Fields

Near-Light Color Photometric Stereo for Mono-Chromatic Non-Lambertian Surfaces

Spike Stream Denoising via Spike Camera Simulation

NTIRE 2025 Challenge on Event-Based Image Deblurring: Methods and Results

No Redundancy, No Stall: Lightweight Streaming 3D Gaussian Splatting for Real-time Rendering

Pano-NeRF: Synthesizing High Dynamic Range Novel Views with Geometry from Sparse Low Dynamic Range Panoramic Images

L-C4: Language-Based Video Colorization for Creative and Consistent Color

Real-Time Intermediate Flow Estimation for Video Frame Interpolation

High-Speed Full-Color HDR Imaging via Unwrapping Modulo-Encoded Spike Streams

Lighting-grounded Video Generation with Renderer-based Agent Reasoning

TRIDENT: A Trimodal Cascade Generative Framework for Drug and RNA-Conditioned Cellular Morphology Synthesis

Technique Report of CVPR 2024 PBDL Challenges

Beyond Dropout: Feature Map Distortion to Regularize Deep Neural Networks