Projects Active Vision Lab
image test

Oxford Active Vision Laboratory Code

The Oxford Active Vision Laboratory Code Repository is a loose collection of computer vision library and applications provided by the Oxford Active Vision Lab from the Department of Engineering Science, University of Oxford.

Recent Projects

NINormal: Neighbourhood-Insensitive Point Cloud Normal Estimation Network
[project] [code] [paper] [video] [supp]

GroSS: Group-Size Series Decomposition for Grouped Architecture Search
[project] [code] [paper]

Finding Non-Uniform Quantization Schemes using Multi-Task Gaussian Processes
[project] [code] [paper]

ANCNet: Correspondence Networks with Adaptive Neighbourhood Consensus
[project] [code] [paper]

DualRC-Net: Dual-Resolution Correspondence Networks
[project] [paper]

Instance Segmentation of LiDAR Point Clouds
[code] [paper] [author]

GANet: Guided Aggregation Net for End-to-end Stereo matching
[code] [paper] [author]

DSConv: Efficient Convolution Operator
[project] [code] [paper]

RelocNet: Continuous Metric Learning Relocalisation Using Neural Nets
[project] [paper]

BNInterpolation: Interpolating Convolutional Neural Networks Using Batch Normalization
[project] [code]

InfiniTAM: A multi-platform framework for real-time, large-scale depth fusion and tracking
[project] [code]

gSLICr: SLIC superpixels at over 250Hz
[project] [code] [paper]

Older Projects

PTAM: Parallel Tracking and Mapping

LibISR: Implicit Shape Representation for 3D Tracking and Reconstruction
[project] [code]

fastHOG: A real-time GPU implementation of HOG
[Tech Report] [single GPU version] [multi GPU version] [git (by Ashwin Nanjappa)] [ubuntu build guide]

PWP3D: Real-time Segmentation and Tracking of 3D Objects
[Paper] [single object and view, VS2008 32bit] [multiple objects and views, VS2010 64bit] [git (by Lu Ma)]

Dense Reconstruction Using 3D Object Shape Priors
[Paper] [Code]