258 citations · 829 across the 41 of their papers we have counts for
43 papers
Matching 2D Images in 3D: Metric Relative Pose from Metric Correspondences
Axel Barroso-Laguna, Sowmya Munukutla, Victor Adrian Prisacariu +1
Given two images, we can estimate the relative camera pose between them by establishing image-to-image correspondences. Usually, correspondences are 2D-to-2D and the pose we estima…
Semi-weakly-supervised neural network training for medical image registration
Yiwen Li, Yunguan Fu, Iani J. M. B. Gayo +11
For training registration networks, weak supervision from segmented corresponding regions-of-interest (ROIs) have been proven effective for (a) supplementing unsupervised methods,…
PoRF: Pose Residual Field for Accurate Neural Surface Reconstruction
Jia-Wang Bian, Wenjing Bian, Victor Adrian Prisacariu +1
Neural surface reconstruction is sensitive to the camera pose noise, even if state-of-the-art pose estimators like COLMAP or ARKit are used. More importantly, existing Pose-NeRF jo…
ABC Easy as 123: A Blind Counter for Exemplar-Free Multi-Class Class-agnostic Counting
Michael A. Hobley, Victor A. Prisacariu
Class-agnostic counting methods enumerate objects of an arbitrary class, providing tremendous utility in many fields. Prior works have limited usefulness as they require either a s…
Cos R-CNN for Online Few-shot Object Detection
Gratianus Wesley Putra Data, Henry Howard-Jenkins, David Murray +1
We propose Cos R-CNN, a simple exemplar-based R-CNN formulation that is designed for online few-shot object detection. That is, it is able to localise and classify novel object cat…
Two-View Geometry Scoring Without Correspondences
Axel Barroso-Laguna, Eric Brachmann, Victor Adrian Prisacariu +2
Camera pose estimation for two-view geometry traditionally relies on RANSAC. Normally, a multitude of image correspondences leads to a pool of proposed hypotheses, which are then s…
DMS: Differentiable Mean Shift for Dataset Agnostic Task Specific Clustering Using Side Information
Michael A. Hobley, Victor A. Prisacariu
We present a novel approach, in which we learn to cluster data directly from side information, in the form of a small set of pairwise examples. Unlike previous methods, with or wit…
Accelerated Coordinate Encoding: Learning to Relocalize in Minutes using RGB and Poses
Eric Brachmann, Tommaso Cavallari, Victor Adrian Prisacariu
Learning-based visual relocalizers exhibit leading pose accuracy, but require hours or days of training. Since training needs to happen on each new scene again, long training times…
Contextualising Implicit Representations for Semantic Tasks
Theo W. Costain, Kejie Li, Victor A. Prisacariu
Prior works have demonstrated that implicit representations trained only for reconstruction tasks typically generate encodings that are not useful for semantic tasks. In this work,…
SimSC: A Simple Framework for Semantic Correspondence with Temperature Learning
Xinghui Li, Kai Han, Xingchen Wan +1
We propose SimSC, a remarkably simple framework, to address the problem of semantic matching only based on the feature backbone. We discover that when fine-tuning ImageNet pre-trai…
MobileBrick: Building LEGO for 3D Reconstruction on Mobile Devices
Kejie Li, Jia-Wang Bian, Robert Castle +2
High-quality 3D ground-truth shapes are critical for 3D object reconstruction evaluation. However, it is difficult to create a replica of an object in reality, and even 3D reconstr…
Neural Refinement for Absolute Pose Regression with Feature Synthesis
Shuai Chen, Yash Bhalgat, Xinghui Li +4
Absolute Pose Regression (APR) methods use deep neural networks to directly regress camera poses from RGB images. However, the predominant APR architectures only rely on 2D operati…
NoPe-NeRF: Optimising Neural Radiance Field with No Pose Prior
Wenjing Bian, Zirui Wang, Kejie Li +2
Training a Neural Radiance Field (NeRF) without pre-computed camera poses is challenging. Recent advances in this direction demonstrate the possibility of jointly optimising a NeRF…
Approximating Continuous Convolutions for Deep Network Compression
Theo W. Costain, Victor Adrian Prisacariu
We present ApproxConv, a novel method for compressing the layers of a convolutional neural network. Reframing conventional discrete convolution as continuous convolution of paramet…
Map-free Visual Relocalization: Metric Pose Relative to a Single Image
Eduardo Arnold, Jamie Wynn, Sara Vicente +5
Can we relocalize in a scene represented by a single reference image? Standard visual relocalization requires hundreds of images and scale calibration to build a scene-specific 3D…
Prototypical few-shot segmentation for cross-institution male pelvic structures with spatial registration
Yiwen Li, Yunguan Fu, Iani Gayo +12
The prowess that makes few-shot learning desirable in medical image analysis is the efficient use of the support image data, which are labelled to classify or segment new classes,…
SimpleRecon: 3D Reconstruction Without 3D Convolutions
Mohamed Sayed, John Gibson, Jamie Watson +3
Traditionally, 3D indoor scene reconstruction from posed images happens in two phases: per-image depth estimation, followed by depth merging and surface reconstruction. Recently, a…
Learning to Count Anything: Reference-less Class-agnostic Counting with Weak Supervision
Michael Hobley, Victor Prisacariu
Current class-agnostic counting methods can generalise to unseen classes but usually require reference images to define the type of object to be counted, as well as instance annota…
BNV-Fusion: Dense 3D Reconstruction using Bi-level Neural Volume Fusion
Kejie Li, Yansong Tang, Victor Adrian Prisacariu +1
Dense 3D reconstruction from a stream of depth images is the key to many mixed reality and robotic applications. Although methods based on Truncated Signed Distance Function (TSDF)…
DFNet: Enhance Absolute Pose Regression with Direct Feature Matching
Shuai Chen, Xinghui Li, Zirui Wang +1
We introduce a camera relocalization pipeline that combines absolute pose regression (APR) and direct feature matching. By incorporating exposure-adaptive novel view synthesis, our…
Few-shot image segmentation for cross-institution male pelvic organs using registration-assisted prototypical learning
Yiwen Li, Yunguan Fu, Qianye Yang +6
The ability to adapt medical image segmentation networks for a novel class such as an unseen anatomical or pathological structure, when only a few labelled examples of this class a…
Few-shot Semantic Segmentation with Self-supervision from Pseudo-classes
Yiwen Li, Gratianus Wesley Putra Data, Yunguan Fu +2
Despite the success of deep learning methods for semantic segmentation, few-shot semantic segmentation remains a challenging task due to the limited training data and the generalis…
Ray-ONet: Efficient 3D Reconstruction From A Single RGB Image
Wenjing Bian, Zirui Wang, Kejie Li +1
We propose Ray-ONet to reconstruct detailed 3D models from monocular images efficiently. By predicting a series of occupancy probabilities along a ray that is back-projected from a…
The Temporal Opportunist: Self-Supervised Multi-Frame Monocular Depth
Jamie Watson, Oisin Mac Aodha, Victor Prisacariu +2
Self-supervised monocular depth estimation networks are trained to predict scene depth using nearby frames as a supervision signal during training. However, for many applications,…