Search Results for author:
Found 52 papers, 16 papers with code
Denoising Monte Carlo Renders With Diffusion Models
Learned methods for restoring low fidelity renders are highly developed, because suppressing render noise means one can save compute and use fast renders with few rays per pixel.
ACDG-VTON: Accurate and Contained Diffusion Generation for Virtual Try-On
Virtual Try-on (VTON) involves generating images of a person wearing selected garments.
HarmBench: A Standardized Evaluation Framework for Automated Red Teaming and Robust Refusal
Automated red teaming holds substantial promise for uncovering and mitigating the risks associated with the malicious use of large language models (LLMs), yet the field lacks a standardized evaluation framework to rigorously assess new methods.
Preserving Image Properties Through Initializations in Diffusion Models
The usual practice of training the denoiser with a very noisy image and starting inference with a sample of pure noise leads to inconsistent generated images during inference.
Improving Equivariance in State-of-the-Art Supervised Depth and Normal Predictors
Dense depth and surface normal predictors should possess the equivariant property to cropping-and-resizing -- cropping the input image should result in cropping the same output image.
Convex Decomposition of Indoor Scenes
Our method uses a learned regression procedure to parse a scene into a fixed number of convexes from RGBD input, and can optionally accept segmentations to improve the decomposition.
Blocks2World: Controlling Realistic Scenes with Editable Primitives
We present Blocks2World, a novel method for 3D scene rendering and editing that leverages a two-step process: convex decomposition of images and conditioned synthesis.
Applying a Color Palette with Local Control using Diffusion Models
We demonstrate two novel editing procedures in the context of fantasy card art.
SoK: Privacy-Preserving Data Synthesis
As the prevalence of data analysis grows, safeguarding data privacy has become a paramount concern.
UrbanIR: Large-Scale Urban Scene Inverse Rendering from a Single Video
UrbanIR uses a novel loss to make very good estimates of shadow volumes in the original scene.
Wearing the Same Outfit in Different Ways -- A Controllable Virtual Try-on Method
However, no current method can both control how the garment is worn -- including tuck or untuck, opened or closed, high or low on the waist, etc.. -- and generate realistic images that accurately preserve the properties of the original garment.
ClimateNeRF: Extreme Weather Synthesis in Neural Radiance Field
Physical simulations produce excellent predictions of weather effects.
How Would The Viewer Feel? Estimating Wellbeing From Video Scenarios
In experiments, we show how video models that are primarily trained to recognize actions and find contours of objects can be repurposed to understand human preferences and the emotional content of videos.
How to Steer Your Adversary: Targeted and Efficient Model Stealing Defenses with Gradient Redirection
To meet these challenges, we present a new approach to model stealing defenses called gradient redirection.
Long Scale Error Control in Low Light Image and Video Enhancement Using Equivariance
In this paper, we show that respecting equivariance -- the color of a restored pixel should be the same, however the image is cropped -- produces real improvements over the state of the art for restoration.
Towards Robust Low Light Image Enhancement
The images are dark because they are taken in dim environments.
DIVeR: Real-time and Accurate Neural Radiance Fields with Deterministic Integration for Volume Rendering
DIVeR's representation is a voxel based field of features.
StyleGAN of All Trades: Image Manipulation with Only Pretrained StyleGAN
Recently, StyleGAN has enabled various image manipulation and editing tasks thanks to the high-quality generation and the disentangled latent space.
On the Importance of Firth Bias Reduction in Few-Shot Classification
In this work, we demonstrate the effectiveness of Firth bias reduction in few-shot classification.
Controlled GAN-Based Creature Synthesis via a Challenging Game Art Dataset -- Addressing the Noise-Latent Trade-Off
While noise inputs to StyleGAN2 are essential for good synthesis, we find that coarse-scale noise interferes with latent variables on this dataset because both control long-scale image effects.
LSD-StructureNet: Modeling Levels of Structural Detail in 3D Part Hierarchies
Generative models for 3D shapes represented by hierarchies of parts can generate realistic and diverse sets of outputs.
Retrieve in Style: Unsupervised Facial Feature Transfer and Retrieval
We present Retrieve in Style (RIS), an unsupervised framework for facial feature transfer and retrieval on real images.
GANs N' Roses: Stable, Controllable, Diverse Image to Image Translation (works for videos too!)
This adversarial loss guarantees the map is diverse -- a very wide range of anime can be produced from a single content code.
Ranked #1 on
Image-to-Image Translation
on selfie2anime
Toward Accurate and Realistic Virtual Try-on Through Shape Matching and Multiple Warps
However, obtaining a realistic image is challenging because the kinematics of garments is complex and because outline, texture, and shading cues in the image reveal errors to human viewers.
Exposing and Correcting the Gender Bias in Image Captioning Datasets and Models
Interestingly, the predictions by this model on images with no humans, are also visibly different from the one trained on gendered captions.
Effectively Unbiased FID and Inception Score and where to find them
In turn, this effectively bias-free estimate requires good estimates of scores with a finite number of samples.
Improving Style Transfer with Calibrated Metrics
Style transfer methods produce a transferred image which is a rendering of a content image in the manner of a style image.
Counterfactual Depth from a Single RGB Image
We describe a method that predicts, from a single RGB image, a depth map that describes the scene when a masked object is removed - we call this "counterfactual depth" that models hidden scene geometry together with the observations.
Coherent and Controllable Outfit Generation
A user study suggests that people understand the match between the queries and the outfits produced by our method.
Why do These Match? Explaining the Behavior of Image Similarity Models
Explaining a deep learning model can help users understand its behavior and allow researchers to discern its shortcomings.
Max-Sliced Wasserstein Distance and its use for GANs
Generative adversarial nets (GANs) and variational auto-encoders have significantly improved our distribution modeling capabilities, showing promise for dataset augmentation, image-to-image translation and feature learning.
Structural Consistency and Controllability for Diverse Colorization
Colorizing a given gray-level image is an important task in the media and advertising industry.
An Approximate Shading Model with Detail Decomposition for Object Relighting
We present an object relighting system that allows an artist to select an object from an image and insert it into a target scene.
Learning Type-Aware Embeddings for Fashion Compatibility
Outfits in online fashion data are composed of items of many different types (e. g. top, bottom, shoes) that share some stylistic relationship with one another.
Detecting Anomalous Faces with 'No Peeking' Autoencoders
We describe a method for detecting an anomalous face image that meets these requirements.
Standard detectors aren't (currently) fooled by physical adversarial stop signs
Finally, an adversarial pattern on a physical object that could fool a detector would have to be adversarial in the face of a wide family of parametric distortions (scale; view angle; box shift inside the detector; illumination; and so on).
NO Need to Worry about Adversarial Examples in Object Detection in Autonomous Vehicles
Instead, a trained neural network classifies most of the pictures taken from different distances and angles of a perturbed image correctly.
SafetyNet: Detecting and Rejecting Adversarial Examples Robustly
SceneProof applies to images captured with depth maps (RGBD images) and checks if a pair of image and depth map is consistent.
Learning Diverse Image Colorization
Finally, we build a conditional model for the multi-modal distribution between grey-level image and the color field embeddings.
Authoring image decompositions with generative models
Instead, we can learn to decompose an image into layers that are "like this" by authoring generative models for each layer using proxy examples that capture the Platonic ideal (Mondrian images for albedo; rendered 3D primitives for shading; material swatches for shading detail).
A Visual Representation for Editing Face Images
First, it allows a user to directly manipulate various illumination.
CDVAE: Co-embedding Deep Variational Auto Encoder for Conditional Variational Generation
Such a model is difficult to train, because we do not usually have training data containing many different shadings for the same image.
Learning to Localize Little Landmarks
We describe a method to find such landmarks by finding a sequence of latent landmarks, each with a prediction model.
Swapout: Learning an ensemble of deep architectures
When viewed as a regularization method swapout not only inhibits co-adaptation of units in a layer, similar to dropout, but also across network layers.
Learning Large-Scale Automatic Image Colorization
The coefficients of the objective function are conditioned on image features, using a random forest.
An Approximate Shading Model for Object Relighting
With this object model, we build an object relighting system that allows an artist to select an object from an image and insert it into a 3D scene.
Sparse Depth Super Resolution
We describe a method to produce detailed high resolution depth maps from aggressively subsampled depth measurements.
Learning a Sequential Search for Landmarks
We propose a general method to find landmarks in images of objects using both appearance and spatial context.
Fast Template Evaluation with Vector Quantization
Our procedure allows speed and accuracy to be traded off in two ways: by choosing the number of Vector Quantization levels, and by choosing to rescore windows or not.
Non-parametric Filtering for Geometric Detail Extraction and Material Representation
Geometric detail is a universal phenomenon in real world objects.
