Search Results for author:
Found 100 papers, 37 papers with code
Fast Convolutional Sparse Coding
Sparse coding has become an increasingly popular method in learning and vision for a variety of classification, reconstruction and coding tasks.
Learning detectors quickly using structured covariance matrices
Computer vision is increasingly becoming interested in the rapid estimation of object detectors.
Correlation Filters with Limited Boundaries
In this paper, we propose a novel approach to correlation filter estimation that: (i) takes advantage of inherent computational redundancies in the frequency domain, and (ii) dramatically reduces boundary effects.
Complex Non-Rigid Motion 3D Reconstruction by Union of Subspaces
The task of estimating complex non-rigid 3D motion through a monocular camera is of increasing interest to the wider scientific community.
Optimization Methods for Convolutional Sparse Coding
We discuss a range of optimization methods for solving the convolutional sparse coding objective, and the properties that make each method suitable for different applications.
Why do linear SVMs trained on HOG features perform so well?
Linear Support Vector Machines trained on HOG features are now a de facto standard across many visual perception tasks.
Regression-Based Image Alignment for General Object Categories
Gradient-descent methods have exhibited fast and reliable performance for image alignment in the facial domain, but have largely been ignored by the broader vision community.
Dense Semantic Correspondence where Every Pixel is a Classifier
Determining dense semantic correspondences across objects and scenes is a difficult problem that underpins many higher-level computer vision algorithms.
Learning Temporal Alignment Uncertainty for Efficient Event Detection
A popular approach in this regard is to represent a sequence using a bag of words (BOW) representation due to its: (i) fixed dimensionality irrespective of the sequence length, and (ii) its ability to compactly model the statistics in the sequence.
Bit-Planes: Dense Subpixel Alignment of Binary Descriptors
Binary descriptors have been instrumental in the recent evolution of computationally efficient sparse image alignment algorithms.
The Conditional Lucas & Kanade Algorithm
In this paper, we present a new approach, referred to as the Conditional LK algorithm, which: (i) directly learns linear models that predict geometric displacement as a function of appearance, and (ii) employs a novel strategy for ensuring that the generative pixel independence assumption can still be taken advantage of.
Direct Visual Odometry using Bit-Planes
Our approach utilizes an efficient binary descriptor, which we call Bit-Planes, and show how it can be used in the gradient-based optimization required by direct methods.
Prior-Less Compressible Structure From Motion
Many non-rigid 3D structures are not modelled well through a low-rank subspace assumption.
Photometric Bundle Adjustment for Vision-Based SLAM
We propose a novel algorithm for the joint refinement of structure and motion parameters from image data directly without relying on fixed and known correspondences.
Inverse Compositional Spatial Transformer Networks
In this paper, we establish a theoretical connection between the classical Lucas & Kanade (LK) algorithm and the emerging topic of Spatial Transformer Networks (STNs).
Fast, Dense Feature SDM on an iPhone
In this paper, we present our method for enabling dense SDM to run at over 90 FPS on a mobile device.
Learning Background-Aware Correlation Filters for Visual Tracking
Correlation Filters (CFs) have recently demonstrated excellent performance in terms of rapidly tracking objects under challenging photometric and geometric variations.
Need for Speed: A Benchmark for Higher Frame Rate Object Tracking
In this paper, we propose the first higher frame rate video dataset (called Need for Speed - NfS) and benchmark for visual object tracking.
Proxy Templates for Inverse Compositional Photometric Bundle Adjustment
In this paper, we show that an ordinary inverse compositional formulation does not work for warps of this type of parameterization due to ill-conditioning of its partial derivatives.
Deep-LK for Efficient Adaptive Object Tracking
In this paper we present a new approach for efficient regression based object tracking which we refer to as Deep- LK.
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 domain (i. e. events with unknown starting and ending locations) is addressed.
Learning Efficient Point Cloud Generation for Dense 3D Object Reconstruction
Conventional methods of 3D object generative modeling learn volumetric predictions using deep networks with 3D convolutional operations, which are direct analogies to classical 2D ones.
Using Locally Corresponding CAD Models for Dense 3D Reconstructions From a Single Image
A common strategy in dictionary learning to encourage generalization is to allow for linear combinations of dictionary elements.
Rethinking Reprojection: Closing the Loop for Pose-aware ShapeReconstruction from a Single Image
An emerging problem in computer vision is the reconstruction of 3D shape and pose of an object from a single image.
Compact Model Representation for 3D Reconstruction
3D reconstruction from 2D images is a central problem in computer vision.
Learning Policies for Adaptive Tracking with Deep Feature Cascades
Our fundamental insight is to take an adaptive approach, where easy frames are processed with cheap features (such as pixel values), while challenging frames are processed with invariant but expensive deep features.
Rethinking Reprojection: Closing the Loop for Pose-Aware Shape Reconstruction From a Single Image
An emerging problem in computer vision is the reconstruction of 3D shape and pose of an object from a single image.
Object-Centric Photometric Bundle Adjustment with Deep Shape Prior
Reconstructing 3D shapes from a sequence of images has long been a problem of interest in computer vision.
Image2Mesh: A Learning Framework for Single Image 3D Reconstruction
One challenge that remains open in 3D deep learning is how to efficiently represent 3D data to feed deep networks.
Semantic Photometric Bundle Adjustment on Natural Sequences
More recently, excellent results have been attained through the application of photometric bundle adjustment (PBA) methods -- which directly minimize the photometric error across frames.
Learning Depth from Monocular Videos using Direct Methods
The ability to predict depth from a single image - using recent advances in CNNs - is of increasing interest to the vision community.
Take it in your stride: Do we need striding in CNNs?
Since their inception, CNNs have utilized some type of striding operator to reduce the overlap of receptive fields and spatial dimensions.
CNNs are Globally Optimal Given Multi-Layer Support
Stochastic Gradient Descent (SGD) is the central workhorse for training modern CNNs.
ST-GAN: Spatial Transformer Generative Adversarial Networks for Image Compositing
We address the problem of finding realistic geometric corrections to a foreground object such that it appears natural when composited into a background image.
Deep Component Analysis via Alternating Direction Neural Networks
Despite a lack of theoretical understanding, deep neural networks have achieved unparalleled performance in a wide range of applications.
Aligning Across Large Gaps in Time
We present a method of temporally-invariant image registration for outdoor scenes, with invariance across time of day, across seasonal variations, and across decade-long periods, for low- and high-texture scenes.
Deep Convolutional Compressed Sensing for LiDAR Depth Completion
In this paper we consider the problem of estimating a dense depth map from a set of sparse LiDAR points.
Deep Interpretable Non-Rigid Structure from Motion
All current non-rigid structure from motion (NRSfM) algorithms are limited with respect to: (i) the number of images, and (ii) the type of shape variability they can handle.
PointNetLK: Robust & Efficient Point Cloud Registration using PointNet
To date, the successful application of PointNet to point cloud registration has remained elusive.
Photometric Mesh Optimization for Video-Aligned 3D Object Reconstruction
In this paper, we address the problem of 3D object mesh reconstruction from RGB videos.
Web Stereo Video Supervision for Depth Prediction from Dynamic Scenes
We present a fully data-driven method to compute depth from diverse monocular video sequences that contain large amounts of non-rigid objects, e. g., people.
Learning Unsupervised Multi-View Stereopsis via Robust Photometric Consistency
We demonstrate our ability to learn MVS without 3D supervision using a real dataset, and show that each component of our proposed robust loss results in a significant improvement.
Deep Non-Rigid Structure from Motion with Missing Data
Non-Rigid Structure from Motion (NRSfM) refers to the problem of reconstructing cameras and the 3D point cloud of a non-rigid object from an ensemble of images with 2D correspondences.
Deep Non-Rigid Structure from Motion
Current non-rigid structure from motion (NRSfM) algorithms are mainly limited with respect to: (i) the number of images, and (ii) the type of shape variability they can handle.
Distill Knowledge from NRSfM for Weakly Supervised 3D Pose Learning
This alleviates the data bottleneck, which is one of the major concern for supervised methods.
Ranked #21 on
Weakly-supervised 3D Human Pose Estimation
on Human3.6M
PCRNet: Point Cloud Registration Network using PointNet Encoding
PointNet has recently emerged as a popular representation for unstructured point cloud data, allowing application of deep learning to tasks such as object detection, segmentation and shape completion.
Argoverse: 3D Tracking and Forecasting with Rich Maps
In our baseline experiments, we illustrate how detailed map information such as lane direction, driveable area, and ground height improves the accuracy of 3D object tracking and motion forecasting.
One Framework to Register Them All: PointNet Encoding for Point Cloud Alignment
We perform extensive simulation and real-world experiments to validate the efficacy of our approach and compare the performance with state-of-art approaches.
Deep NRSfM++: Towards Unsupervised 2D-3D Lifting in the Wild
The recovery of 3D shape and pose from 2D landmarks stemming from a large ensemble of images can be viewed as a non-rigid structure from motion (NRSfM) problem.
High Accuracy Face Geometry Capture using a Smartphone Video
What's the most accurate 3D model of your face you can obtain while sitting at your desk?
Dataless Model Selection with the Deep Frame Potential
Choosing a deep neural network architecture is a fundamental problem in applications that require balancing performance and parameter efficiency.
When to Use Convolutional Neural Networks for Inverse Problems
In this work we argue that for some types of inverse problems the CNN approximation breaks down leading to poor performance.
PointNetLK Revisited
We address the generalization ability of recent learning-based point cloud registration methods.
Joint Pose and Shape Estimation of Vehicles from LiDAR Data
We address the problem of estimating the pose and shape of vehicles from LiDAR scans, a common problem faced by the autonomous vehicle community.
MaskNet: A Fully-Convolutional Network to Estimate Inlier Points
We demonstrate these improvements on synthetic and real-world datasets.
SDF-SRN: Learning Signed Distance 3D Object Reconstruction from Static Images
Dense 3D object reconstruction from a single image has recently witnessed remarkable advances, but supervising neural networks with ground-truth 3D shapes is impractical due to the laborious process of creating paired image-shape datasets.
3D Object Reconstruction From A Single Image
3D Reconstruction
Scene Flow from Point Clouds with or without Learning
Our proposed objective function can be used with or without learning---as a self-supervisory signal to learn scene flow representations, or as a non-learning-based method in which the scene flow is optimized during runtime.
Architectural Adversarial Robustness: The Case for Deep Pursuit
Despite their unmatched performance, deep neural networks remain susceptible to targeted attacks by nearly imperceptible levels of adversarial noise.
Reframing Neural Networks: Deep Structure in Overcomplete Representations
In comparison to classical shallow representation learning techniques, deep neural networks have achieved superior performance in nearly every application benchmark.
PAUL: Procrustean Autoencoder for Unsupervised Lifting
Recent success in casting Non-rigid Structure from Motion (NRSfM) as an unsupervised deep learning problem has raised fundamental questions about what novelty in NRSfM prior could the deep learning offer.
BARF: Bundle-Adjusting Neural Radiance Fields
In this paper, we propose Bundle-Adjusting Neural Radiance Fields (BARF) for training NeRF from imperfect (or even unknown) camera poses -- the joint problem of learning neural 3D representations and registering camera frames.
Evading the Simplicity Bias: Training a Diverse Set of Models Discovers Solutions with Superior OOD Generalization
The method - the first to evade the simplicity bias - highlights the need for a better understanding and control of inductive biases in deep learning.
Neural Trajectory Fields for Dynamic Novel View Synthesis
Recent approaches to render photorealistic views from a limited set of photographs have pushed the boundaries of our interactions with pictures of static scenes.
On the Bias Against Inductive Biases
Borrowing from the transformer models that revolutionized the field of natural language processing, self-supervised feature learning for visual tasks has also seen state-of-the-art success using these extremely deep, isotropic networks.
Rethinking Positional Encoding
It is well noted that coordinate based MLPs benefit -- in terms of preserving high-frequency information -- through the encoding of coordinate positions as an array of Fourier features.
High Fidelity 3D Reconstructions with Limited Physical Views
Multi-view triangulation is the gold standard for 3D reconstruction from 2D correspondences given known calibration and sufficient views.
Neural Scene Flow Prior
A central innovation here is the inclusion of a neural scene flow prior, which uses the architecture of neural networks as a new type of implicit regularizer.
Ranked #2 on
Self-supervised Scene Flow Estimation
on Argoverse 2
Enabling equivariance for arbitrary Lie groups
Our framework enables the implementation of group convolutions over any finite-dimensional Lie group.
Beyond Periodicity: Towards a Unifying Framework for Activations in Coordinate-MLPs
Coordinate-MLPs are emerging as an effective tool for modeling multidimensional continuous signals, overcoming many drawbacks associated with discrete grid-based approximations.
Learning Positional Embeddings for Coordinate-MLPs
We propose a novel method to enhance the performance of coordinate-MLPs by learning instance-specific positional embeddings.
Neural Prior for Trajectory Estimation
Here, we propose a neural trajectory prior to capture continuous spatio-temporal information without the need for offline data.
On Regularizing Coordinate-MLPs
We show that typical implicit regularization assumptions for deep neural networks (for regression) do not hold for coordinate-MLPs, a family of MLPs that are now ubiquitous in computer vision for representing high-frequency signals.
Long-term Visual Map Sparsification with Heterogeneous GNN
We address the problem of map sparsification for long-term visual localization.
GARF: Gaussian Activated Radiance Fields for High Fidelity Reconstruction and Pose Estimation
Despite Neural Radiance Fields (NeRF) showing compelling results in photorealistic novel views synthesis of real-world scenes, most existing approaches require accurate prior camera poses.
Trading Positional Complexity vs. Deepness in Coordinate Networks
It is well noted that coordinate-based MLPs benefit -- in terms of preserving high-frequency information -- through the encoding of coordinate positions as an array of Fourier features.
How You Start Matters for Generalization
Characterizing the remarkable generalization properties of over-parameterized neural networks remains an open problem.
On Quantizing Implicit Neural Representations
The role of quantization within implicit/coordinate neural networks is still not fully understood.
MBW: Multi-view Bootstrapping in the Wild
Our Multi-view Bootstrapping in the Wild (MBW) approach demonstrates impressive results on standard human datasets, as well as tigers, cheetahs, fish, colobus monkeys, chimpanzees, and flamingos from videos captured casually in a zoo.
3D Reconstruction
Semi-supervised 2D and 3D landmark labeling
+1
On skip connections and normalisation layers in deep optimisation
We introduce a general theoretical framework, designed for the study of gradient optimisation of deep neural networks, that encompasses ubiquitous architecture choices including batch normalisation, weight normalisation and skip connections.
Neural Radiance Field with LiDAR maps
We address outdoor Neural Radiance Fields (NeRF) with LiDAR maps.
On the effectiveness of neural priors in modeling dynamical systems
Modelling dynamical systems is an integral component for understanding the natural world.
Flow supervision for Deformable NeRF
In this paper we present a new method for deformable NeRF that can directly use optical flow as supervision.
Re-Evaluating LiDAR Scene Flow for Autonomous Driving
Popular benchmarks for self-supervised LiDAR scene flow (stereoKITTI, and FlyingThings3D) have unrealistic rates of dynamic motion, unrealistic correspondences, and unrealistic sampling patterns.
Fast Neural Scene Flow
Neural Scene Flow Prior (NSFP) is of significant interest to the vision community due to its inherent robustness to out-of-distribution (OOD) effects and its ability to deal with dense lidar points.
Ranked #4 on
Self-supervised Scene Flow Estimation
on Argoverse 2
Curvature-Aware Training for Coordinate Networks
Coordinate networks are widely used in computer vision due to their ability to represent signals as compressed, continuous entities.
On progressive sharpening, flat minima and generalisation
We present a new approach to understanding the relationship between loss curvature and input-output model behaviour in deep learning.
Robust Point Cloud Processing through Positional Embedding
End-to-end trained per-point embeddings are an essential ingredient of any state-of-the-art 3D point cloud processing such as detection or alignment.
Multi-Body Neural Scene Flow
The test-time optimization of scene flow - using a coordinate network as a neural prior - has gained popularity due to its simplicity, lack of dataset bias, and state-of-the-art performance.
3D-LFM: Lifting Foundation Model
The lifting of 3D structure and camera from 2D landmarks is at the cornerstone of the entire discipline of computer vision.
Ranked #1 on
3D Hand Pose Estimation
on H3WB
Convolutional Initialization for Data-Efficient Vision Transformers
Training vision transformer networks on small datasets poses challenges.
Architectural Strategies for the optimization of Physics-Informed Neural Networks
Physics-informed neural networks (PINNs) offer a promising avenue for tackling both forward and inverse problems in partial differential equations (PDEs) by incorporating deep learning with fundamental physics principles.
Analyzing the Neural Tangent Kernel of Periodically Activated Coordinate Networks
In this paper, we aim to address this gap by providing a theoretical understanding of periodically activated networks through an analysis of their Neural Tangent Kernel (NTK).
A Sampling Theory Perspective on Activations for Implicit Neural Representations
Implicit Neural Representations (INRs) have gained popularity for encoding signals as compact, differentiable entities.
Preconditioners for the Stochastic Training of Implicit Neural Representations
Implicit neural representations have emerged as a powerful technique for encoding complex continuous multidimensional signals as neural networks, enabling a wide range of applications in computer vision, robotics, and geometry.
SeMoLi: What Moves Together Belongs Together
Recent results suggest that heuristic-based clustering methods in conjunction with object trackers can be used to pseudo-label instances of moving objects and use these as supervisory signals to train 3D object detectors in Lidar data without manual supervision.
Fast Kernel Scene Flow
In contrast to current state-of-the-art methods, such as NSFP [25], which employ deep implicit neural functions for modeling scene flow, we present a novel approach that utilizes classical kernel representations.
Sine Activated Low-Rank Matrices for Parameter Efficient Learning
Low-rank decomposition has emerged as a vital tool for enhancing parameter efficiency in neural network architectures, gaining traction across diverse applications in machine learning.
From Activation to Initialization: Scaling Insights for Optimizing Neural Fields
In the realm of computer vision, Neural Fields have gained prominence as a contemporary tool harnessing neural networks for signal representation.
D'OH: Decoder-Only random Hypernetworks for Implicit Neural Representations
We instead present a strategy for the initialization of run-time deep implicit functions for single-instance signals through a Decoder-Only randomly projected Hypernetwork (D'OH).
Structured Initialization for Attention in Vision Transformers
By contrast, convolutional neural networks (CNNs) have an architectural inductive bias enabling them to perform well on such problems.
