Search Results for author:
Found 41 papers, 21 papers with code
Rotationally-Temporally Consistent Novel View Synthesis of Human Performance Video
Second, we introduce a novel loss to explicitly enforce consistency across generated views both in space and in time.
EditSplat: Multi-View Fusion and Attention-Guided Optimization for View-Consistent 3D Scene Editing with 3D Gaussian Splatting
Recent advancements in 3D editing have highlighted the potential of text-driven methods in real-time, user-friendly AR/VR applications.
Sequence Matters: Harnessing Video Models in 3D Super-Resolution
By utilizing VSR models, we ensure a higher degree of spatial consistency and can reference surrounding spatial information, leading to more accurate and detailed reconstructions.
Generative Densification: Learning to Densify Gaussians for High-Fidelity Generalizable 3D Reconstruction
Generalized feed-forward Gaussian models have achieved significant progress in sparse-view 3D reconstruction by leveraging prior knowledge from large multi-view datasets.
PIG: Physics-Informed Gaussians as Adaptive Parametric Mesh Representations
The approximation of Partial Differential Equations (PDEs) using neural networks has seen significant advancements through Physics-Informed Neural Networks (PINNs).
MetaFormer: High-fidelity Metalens Imaging via Aberration Correcting Transformers
We further fabricate a metalens and verify the practicality of MetaFormer by restoring the images captured with the manufactured metalens in the wild.
SelfSplat: Pose-Free and 3D Prior-Free Generalizable 3D Gaussian Splatting
Our model addresses these challenges by effectively integrating explicit 3D representations with self-supervised depth and pose estimation techniques, resulting in reciprocal improvements in both pose accuracy and 3D reconstruction quality.
Compact 3D Gaussian Splatting for Static and Dynamic Radiance Fields
With model compression techniques such as quantization and entropy coding, we consistently show over 25x reduced storage and enhanced rendering speed compared to 3DGS for static scenes, while maintaining the quality of the scene representation.
DiffuseHigh: Training-free Progressive High-Resolution Image Synthesis through Structure Guidance
In this work, we probe the generative ability of diffusion models at higher resolution beyond their original capability and propose a novel progressive approach that fully utilizes generated low-resolution images to guide the generation of higher-resolution images.
Freq-Mip-AA : Frequency Mip Representation for Anti-Aliasing Neural Radiance Fields
The proposed method, FreqMipAA, utilizes scale-specific low-pass filtering (LPF) and learnable frequency masks.
F-3DGS: Factorized Coordinates and Representations for 3D Gaussian Splatting
To mitigate the storage overhead, we propose Factorized 3D Gaussian Splatting (F-3DGS), a novel approach that drastically reduces storage requirements while preserving image quality.
Parameter-Efficient Instance-Adaptive Neural Video Compression
Inspired by the remarkable success of parameter-efficient fine-tuning on large-scale neural network models, we propose to use a lightweight adapter module that can be easily attached to the pretrained NVCs and fine-tuned for test video sequences.
CodecNeRF: Toward Fast Encoding and Decoding, Compact, and High-quality Novel-view Synthesis
Neural Radiance Fields (NeRF) have achieved huge success in effectively capturing and representing 3D objects and scenes.
Separable Physics-informed Neural Networks for Solving the BGK Model of the Boltzmann Equation
In this study, we introduce a method based on Separable Physics-Informed Neural Networks (SPINNs) for effectively solving the BGK model of the Boltzmann equation.
Continuous Memory Representation for Anomaly Detection
To tackle all of the above challenges, we propose CRAD, a novel anomaly detection method for representing normal features within a "continuous" memory, enabled by transforming spatial features into coordinates and mapping them to continuous grids.
Ranked #31 on
Anomaly Detection
on MVTec AD
Mip-Grid: Anti-aliased Grid Representations for Neural Radiance Fields
In this work, we present mip-Grid, a novel approach that integrates anti-aliasing techniques into grid-based representations for radiance fields, mitigating the aliasing artifacts while enjoying fast training time.
Sharp-NeRF: Grid-based Fast Deblurring Neural Radiance Fields Using Sharpness Prior
Especially, defocus blur is quite common in the images when they are normally captured using cameras.
Deblurring 3D Gaussian Splatting
However, it suffers from severe degradation in the rendering quality if the training images are blurry.
Residual Learning in Diffusion Models
This gap results in a residual in the generated images adversely impacting the image quality.
Coordinate-Aware Modulation for Neural Fields
Experimental results demonstrate that CAM enhances the performance of neural representation and improves learning stability across a range of signals.
Compact 3D Gaussian Representation for Radiance Field
On the other hand, 3D Gaussian splatting (3DGS) has recently emerged as an alternative representation that leverages a 3D Gaussisan-based representation and adopts the rasterization pipeline to render the images rather than volumetric rendering, achieving very fast rendering speed and promising image quality.
Ranked #9 on
Novel View Synthesis
on Mip-NeRF 360
Hydra: Multi-head Low-rank Adaptation for Parameter Efficient Fine-tuning
The recent surge in large-scale foundation models has spurred the development of efficient methods for adapting these models to various downstream tasks.
Separable Physics-Informed Neural Networks
Furthermore, we present that SPINN can solve a chaotic (2+1)-d Navier-Stokes equation significantly faster than the best-performing prior method (9 minutes vs 10 hours in a single GPU), maintaining accuracy.
SMPConv: Self-moving Point Representations for Continuous Convolution
This paper suggests an alternative approach to building a continuous convolution without neural networks, resulting in more computationally efficient and improved performance.
Ranked #1 on
Sequential Image Classification
on Sequential MNIST
FFNeRV: Flow-Guided Frame-Wise Neural Representations for Videos
Neural fields, also known as coordinate-based or implicit neural representations, have shown a remarkable capability of representing, generating, and manipulating various forms of signals.
Ranked #2 on
Video Reconstruction
on UVG
Masked Wavelet Representation for Compact Neural Radiance Fields
There have been recent studies on how to reduce these computational inefficiencies by using additional data structures, such as grids or trees.
Separable PINN: Mitigating the Curse of Dimensionality in Physics-Informed Neural Networks
SPINN operates on a per-axis basis instead of point-wise processing in conventional PINNs, decreasing the number of network forward passes.
PIXEL: Physics-Informed Cell Representations for Fast and Accurate PDE Solvers
With the increases in computational power and advances in machine learning, data-driven learning-based methods have gained significant attention in solving PDEs.
Streamable Neural Fields
Neural fields have emerged as a new data representation paradigm and have shown remarkable success in various signal representations.
Neural Residual Flow Fields for Efficient Video Representations
Inspired by standard video compression algorithms, we propose a neural field architecture for representing and compressing videos that deliberately removes data redundancy through the use of motion information across video frames.
IA-MARL: Imputation Assisted Multi-Agent Reinforcement Learning for Missing Training Data
Recently, multi-agent reinforcement learning (MARL) adopts the centralized training with decentralized execution (CTDE) framework that trains agents using the data from all agents at a centralized server while each agent takes an action from its observation.
Unsupervised Doodling and Painting with Improved SPIRAL
We investigate using reinforcement learning agents as generative models of images (extending arXiv:1804. 01118).
Low-Power Computer Vision: Status, Challenges, Opportunities
In addition to mobile phones, many autonomous systems rely on visual data for making decisions and some of these systems have limited energy (such as unmanned aerial vehicles also called drones and mobile robots).
Meta-Curvature
We propose meta-curvature (MC), a framework to learn curvature information for better generalization and fast model adaptation.
2018 Low-Power Image Recognition Challenge
The Low-Power Image Recognition Challenge (LPIRC, https://rebootingcomputing. ieee. org/lpirc) is an annual competition started in 2015.
Meta-Tracker: Fast and Robust Online Adaptation for Visual Object Trackers
The meta learning is driven by the goal of deep networks that can quickly be adapted to robustly model a particular target in future frames.
Transformation-Grounded Image Generation Network for Novel 3D View Synthesis
Instead of taking a 'blank slate' approach, we first explicitly infer the parts of the geometry visible both in the input and novel views and then re-cast the remaining synthesis problem as image completion.
A Dataset for Developing and Benchmarking Active Vision
We present a new public dataset with a focus on simulating robotic vision tasks in everyday indoor environments using real imagery.
Visual Madlibs: Fill in the Blank Description Generation and Question Answering
In this paper, we introduce a new dataset consisting of 360, 001 focused natural language descriptions for 10, 738 images.
Learning to decompose for object detection and instance segmentation
Although deep convolutional neural networks(CNNs) have achieved remarkable results on object detection and segmentation, pre- and post-processing steps such as region proposals and non-maximum suppression(NMS), have been required.
Visual Madlibs: Fill in the blank Image Generation and Question Answering
In this paper, we introduce a new dataset consisting of 360, 001 focused natural language descriptions for 10, 738 images.
