
Beyond Periodicity: Towards a Unifying Framework for Activations in CoordinateMLPs
CoordinateMLPs are emerging as an effective tool for modeling multidime...
read it

Enabling equivariance for arbitrary Lie groups
Although provably robust to translational perturbations, convolutional n...
read it

Neural Scene Flow Prior
Before the deep learning revolution, many perception algorithms were bas...
read it

Rethinking Positional Encoding
It is well noted that coordinate based MLPs benefit greatly – in terms o...
read it

On the Bias Against Inductive Biases
Borrowing from the transformer models that revolutionized the field of n...
read it

Neural Trajectory Fields for Dynamic Novel View Synthesis
Recent approaches to render photorealistic views from a limited set of p...
read it

BARF: BundleAdjusting Neural Radiance Fields
Neural Radiance Fields (NeRF) have recently gained a surge of interest w...
read it

PAUL: Procrustean Autoencoder for Unsupervised Lifting
Recent success in casting Nonrigid Structure from Motion (NRSfM) as an ...
read it

Reframing Neural Networks: Deep Structure in Overcomplete Representations
In comparison to classical shallow representation learning techniques, d...
read it

Architectural Adversarial Robustness: The Case for Deep Pursuit
Despite their unmatched performance, deep neural networks remain suscept...
read it

Scene Flow from Point Clouds with or without Learning
Scene flow is the threedimensional (3D) motion field of a scene. It pro...
read it

SDFSRN: Learning Signed Distance 3D Object Reconstruction from Static Images
Dense 3D object reconstruction from a single image has recently witnesse...
read it

MaskNet: A FullyConvolutional Network to Estimate Inlier Points
Point clouds have grown in importance in the way computers perceive the ...
read it

Joint Pose and Shape Estimation of Vehicles from LiDAR Data
We address the problem of estimating the pose and shape of vehicles from...
read it

Deterministic PointNetLK for Generalized Registration
There has been remarkable progress in the application of deep learning t...
read it

Dataless Model Selection with the Deep Frame Potential
Choosing a deep neural network architecture is a fundamental problem in ...
read it

When to Use Convolutional Neural Networks for Inverse Problems
Reconstruction tasks in computer vision aim fundamentally to recover an ...
read it

High Accuracy Face Geometry Capture using a Smartphone Video
What's the most accurate 3D model of your face you can obtain while sitt...
read it

Deep NRSfM++: Towards 3D Reconstruction in the Wild
The recovery of 3D shape and pose solely from 2D landmarks stemming from...
read it

One Framework to Register Them All: PointNet Encoding for Point Cloud Alignment
PointNet has recently emerged as a popular representation for unstructur...
read it

Argoverse: 3D Tracking and Forecasting with Rich Maps
We present Argoverse – two datasets designed to support autonomous vehic...
read it

PCRNet: Point Cloud Registration Network using PointNet Encoding
PointNet has recently emerged as a popular representation for unstructur...
read it

Distill Knowledge from NRSfM for Weakly Supervised 3D Pose Learning
We propose to learn a 3D pose estimator by distilling knowledge from Non...
read it

Deep NonRigid Structure from Motion
NonRigid Structure from Motion (NRSfM) refers to the problem of reconst...
read it

Learning Unsupervised MultiView Stereopsis via Robust Photometric Consistency
We present a learning based approach for multiview stereopsis (MVS). Wh...
read it

Web Stereo Video Supervision for Depth Prediction from Dynamic Scenes
We present a fully datadriven method to compute depth from diverse mono...
read it

Photometric Mesh Optimization for VideoAligned 3D Object Reconstruction
In this paper, we address the problem of 3D object mesh reconstruction f...
read it

PointNetLK: Robust & Efficient Point Cloud Registration using PointNet
PointNet has revolutionized how we think about representing point clouds...
read it

Deep Interpretable NonRigid Structure from Motion
All current nonrigid structure from motion (NRSfM) algorithms are limit...
read it

Deep Convolutional Compressed Sensing for LiDAR Depth Completion
In this paper we consider the problem of estimating a dense depth map fr...
read it

Aligning Across Large Gaps in Time
We present a method of temporallyinvariant image registration for outdo...
read it

Deep Component Analysis via Alternating Direction Neural Networks
Despite a lack of theoretical understanding, deep neural networks have a...
read it

STGAN: Spatial Transformer Generative Adversarial Networks for Image Compositing
We address the problem of finding realistic geometric corrections to a f...
read it

Take it in your stride: Do we need striding in CNNs?
Since their inception, CNNs have utilized some type of striding operator...
read it

CNNs are Globally Optimal Given MultiLayer Support
Stochastic Gradient Descent (SGD) is the central workhorse for training ...
read it

Learning Depth from Monocular Videos using Direct Methods
The ability to predict depth from a single image  using recent advances...
read it

Semantic Photometric Bundle Adjustment on Natural Sequences
The problem of obtaining dense reconstruction of an object in a natural ...
read it

Image2Mesh: A Learning Framework for Single Image 3D Reconstruction
One challenge that remains open in 3D deep learning is how to efficientl...
read it

ObjectCentric Photometric Bundle Adjustment with Deep Shape Prior
Reconstructing 3D shapes from a sequence of images has long been a probl...
read it

Learning Policies for Adaptive Tracking with Deep Feature Cascades
Visual object tracking is a fundamental and timecritical vision task. R...
read it

Compact Model Representation for 3D Reconstruction
3D reconstruction from 2D images is a central problem in computer vision...
read it

Rethinking Reprojection: Closing the Loop for Poseaware ShapeReconstruction from a Single Image
An emerging problem in computer vision is the reconstruction of 3D shape...
read it

Learning Efficient Point Cloud Generation for Dense 3D Object Reconstruction
Conventional methods of 3D object generative modeling learn volumetric p...
read it

Joint Max Margin and Semantic Features for Continuous Event Detection in Complex Scenes
In this paper the problem of complex event detection in the continuous d...
read it

DeepLK for Efficient Adaptive Object Tracking
In this paper we present a new approach for efficient regression based o...
read it

Need for Speed: A Benchmark for Higher Frame Rate Object Tracking
In this paper, we propose the first higher frame rate video dataset (cal...
read it

Learning BackgroundAware Correlation Filters for Visual Tracking
Correlation Filters (CFs) have recently demonstrated excellent performan...
read it

Fast, Dense Feature SDM on an iPhone
In this paper, we present our method for enabling dense SDM to run at ov...
read it

Inverse Compositional Spatial Transformer Networks
In this paper, we establish a theoretical connection between the classic...
read it

Photometric Bundle Adjustment for VisionBased SLAM
We propose a novel algorithm for the joint refinement of structure and m...
read it
Simon Lucey
is this you? claim profile
Associate Research Professor, Robotics Institute at Carnegie Mellon University