Search Results for author:
Found 38 papers, 11 papers with code
MVPSNet: Fast Generalizable Multi-view Photometric Stereo
We propose a fast and generalizable solution to Multi-view Photometric Stereo (MVPS), called MVPSNet.
Supervision Interpolation via LossMix: Generalizing Mixup for Object Detection and Beyond
The success of data mixing augmentations in image classification tasks has been well-received.
A Surface-normal Based Neural Framework for Colonoscopy Reconstruction
Reconstructing a 3D surface from colonoscopy video is challenging due to illumination and reflectivity variation in the video frame that can cause defective shape predictions.
A Practical Stereo Depth System for Smart Glasses
We present the design of a productionized end-to-end stereo depth sensing system that does pre-processing, online stereo rectification, and stereo depth estimation with a fallback to monocular depth estimation when rectification is unreliable.
Toward Edge-Efficient Dense Predictions with Synergistic Multi-Task Neural Architecture Search
Our first key insight is that MultiTask Learning (MTL) and hardware-aware Neural Architecture Search (NAS) can work in synergy to greatly benefit on-device Dense Predictions (DP).
Ranked #111 on
Semantic Segmentation
on NYU Depth v2
Hardware Aware Neural Architecture Search
Neural Architecture Search
+2
Leveraging Disentangled Representations to Improve Vision-Based Keystroke Inference Attacks Under Low Data
We evaluate our method on real-life data using a variety of metrics to quantify the amount of information an attacker is able to recover.
VPFusion: Joint 3D Volume and Pixel-Aligned Feature Fusion for Single and Multi-view 3D Reconstruction
VPFusion attains high-quality reconstruction using both - 3D feature volume to capture 3D-structure-aware context, and pixel-aligned image features to capture fine local detail.
ColDE: A Depth Estimation Framework for Colonoscopy Reconstruction
Aiming to fundamentally improve the depth estimation quality for colonoscopy 3D reconstruction, in this work we have designed a set of training losses to deal with the special challenges of colonoscopy data.
Disentangling style and content for low resource video domain adaptation: a case study on keystroke inference attacks
We evaluate our method on real-life data using a variety of metrics to quantify the amount of information an attacker is able to recover.
Any-Width Networks
To address these constraints, we propose the Any-Width Network (AWN), an adjustable-width CNN architecture and associated training routine that allow for fine-grained control over speed and accuracy during inference.
Revisiting the Threat Space for Vision-based Keystroke Inference Attacks
This indicates that these models are able to learn rich, meaningful representations from our synthetic data and that training on the synthetic data can help overcome the issue of having small, real-life datasets for vision-based key stroke inference attacks.
One Shot 3D Photography
3D photos are static in time, like traditional photos, but are displayed with interactive parallax on mobile or desktop screens, as well as on Virtual Reality devices, where viewing it also includes stereo.
Reducing Drift in Structure From Motion Using Extended Features
Low-frequency long-range errors (drift) are an endemic problem in 3D structure from motion, and can often hamper reasonable reconstructions of the scene.
Tangent Images for Mitigating Spherical Distortion
In this work, we propose "tangent images," a spherical image representation that facilitates transferable and scalable $360^\circ$ computer vision.
ViewSynth: Learning Local Features from Depth using View Synthesis
By learning view synthesis, we explicitly encourage the feature extractor to encode information about not only the visible, but also the occluded parts of the scene.
Mapped Convolutions
We present a versatile formulation of the convolution operation that we term a "mapped convolution."
Convolutions on Spherical Images
Applying convolutional neural networks to spherical images requires particular considerations.
Recurrent Neural Network for (Un-)supervised Learning of Monocular VideoVisual Odometry and Depth
Deep learning-based, single-view depth estimation methods have recently shown highly promising results.
Deep Blending for Free-Viewpoint Image-Based-Rendering
We present a new deep learning approach to blending for IBR, in which we use held-out real image data to learn blending weights to combine input photo contributions.
Augmenting Crowd-Sourced 3D Reconstructions Using Semantic Detections
Image-based 3D reconstruction for Internet photo collections has become a robust technology to produce impressive virtual representations of real-world scenes.
Rolling Shutter and Radial Distortion Are Features for High Frame Rate Multi-Camera Tracking
In this paper, we introduce a novel multi-camera tracking approach that for the first time jointly leverages the information introduced by rolling shutter and radial distortion as a feature to achieve superior performance with respect to high-frequency camera pose estimation.
Recurrent Neural Network for Learning DenseDepth and Ego-Motion from Video
Our method produces superior results to the state-of-the-art learning-based, single- or two-view depth estimation methods on both indoor and outdoor benchmark datasets.
The Domain Transform Solver
We present a framework for edge-aware optimization that is an order of magnitude faster than the state of the art while having comparable performance.
Learned Contextual Feature Reweighting for Image Geo-Localization
We address the problem of large scale image geo-localization where the location of an image is estimated by identifying geo-tagged reference images depicting the same place.
The Misty Three Point Algorithm for Relative Pose
There is a significant interest in scene reconstruction from underwater images given its utility for oceanic research and for recreational image manipulation.
Structure-From-Motion Revisited
Incremental Structure-from-Motion is a prevalent strategy for 3D reconstruction from unordered image collections.
Ranked #13 on
Point Clouds
on Tanks and Temples
From Dusk Till Dawn: Modeling in the Dark
We present an algorithm that leverages the appearance variety to obtain more complete and accurate scene geometry along with consistent multi-illumination appearance information.
Self-expressive Dictionary Learning for Dynamic 3D Reconstruction
Given the smooth motion of dynamic objects, we observe any element in the dictionary can be well approximated by a sparse linear combination of other elements in the same dictionary (i. e. self-expression).
Minimal Solvers for 3D Geometry From Satellite Imagery
We propose two novel minimal solvers which advance the state of the art in satellite imagery processing.
Predicting Good Features for Image Geo-Localization Using Per-Bundle VLAD
We address the problem of recognizing a place depicted in a query image by using a large database of geo-tagged images at a city-scale.
Synthesizing Illumination Mosaics From Internet Photo-Collections
We propose a framework for the automatic creation of time-lapse mosaics of a given scene.
Sparse Dynamic 3D Reconstruction From Unsynchronized Videos
We target the sparse 3D reconstruction of dynamic objects observed by multiple unsynchronized video cameras with unknown temporal overlap.
From Single Image Query to Detailed 3D Reconstruction
Structure-from-Motion for unordered image collections has significantly advanced in scale over the last decade.
Reconstructing the World* in Six Days *(As Captured by the Yahoo 100 Million Image Dataset)
We propose a novel, large-scale, structure-from-motion framework that advances the state of the art in data scalability from city-scale modeling (millions of images) to world-scale modeling (several tens of millions of images) using just a single computer.
Adaptive Eye-Camera Calibration for Head-Worn Devices
Our calibration scheme allows a head-worn device to calculate a locally optimal eye-device transformation on demand by computing an optimal model from a local window of previous frames.
PAIGE: PAirwise Image Geometry Encoding for Improved Efficiency in Structure-From-Motion
Based on the insights of this evaluation, we propose a learning-based approach, the PAirwise Image Geometry Encoding (PAIGE), to efficiently identify image pairs with scene overlap without the need to perform exhaustive putative matching and geometric verification.
PatchMatch Based Joint View Selection and Depthmap Estimation
We propose a multi-view depthmap estimation approach aimed at adaptively ascertaining the pixel level data associations between a reference image and all the elements of a source image set.
Stereo under Sequential Optimal Sampling: A Statistical Analysis Framework for Search Space Reduction
We develop a sequential optimal sampling framework for stereo disparity estimation by adapting the Sequential Probability Ratio Test (SPRT) model.
