Proceedings of the British Machine Vision Conference 2016
Editors: Richard C. Wilson, Edwin R. Hancock and William A. P. Smith BMVA Press, September 2016, ISBN 1-901725-59-6
Book of Abstracts (PDF, 206 pages, 172M)Tutorial
Keynote 1
Segmentation
- Local Shape Transfer for Image Co-segmentation
- SMURFS: Superpixels from Multi-scale Refinement of Super-regions
- Biologically plausible boundary detection
- Bio-inspired Collision Detector with Enhanced Selectivity for Ground Robotic Vision System
- A Deep Primal-Dual Network for Guided Depth Super-Resolution
- Towards Deep Style Transfer: A Content-Aware Perspective
- Real-Time Intensity-Image Reconstruction for Event Cameras Using Manifold Regularisation
Posters 1
- Multi-view Multi-illuminant Intrinsic Dataset
- Accurate Closed-form Estimation of Local Affine Transformations Consistent with the Epipolar Geometry
- Recognition of Transitional Action for Short-Term Action Prediction using Discriminative Temporal CNN Feature
- Multi-H: Efficient recovery of tangent planes in stereo images
- Jointly Learning Non-negative Projection and Dictionary with Discriminative Graph Constraints for Classification
- A MultiPath Network for Object Detection
- Fast Eigen Matching
- Recoding Color Transfer as A Color Homography
- Pose-Robust 3D Facial Landmark Estimation from a Single 2D Image
- Bottom-up Instance Segmentation using Deep Higher-Order CRFs
- Horizon Lines in the Wild
- An Octree-Based Approach towards Efficient Variational Range Data Fusion
- Finsler Geodesics Evolution Model for Region based Active Contours
- Patch Based Confidence Prediction for Dense Disparity Map
- Boosted Convolutional Neural Networks
- Material-Specific Chromaticity Priors
- Play and Learn: Using Video Games to Train Computer Vision Models
- SDF-TAR: Parallel Tracking and Refinement in RGB-D Data using Volumetric Registration
- Probabilistic Semi-Supervised Multi-Modal Hashing
- Learning to Invert Local Binary Patterns
- Probabilistic Compositional Active Basis Models for Robust Pattern Recognition
- Loglet SIFT for Part Description in Deformable Part Models: Application to Face Alignment
- Dictionary Replacement for Single Image Restoration of 3D Scenes.
- Poisson Noise Removal for Image Demosaicing
- Factorized Binary Codes for Large-Scale Nearest Neighbor Search
- Edge Enhanced Direct Visual Odometry
- Optimised photometric stereo via non-convex variational minimisation
- U-shaped Networks for Shape from Light Field
- Using Shading and a 3D Template to Reconstruct Complex Surface Deformations
- Physics 101: Learning Physical Object Properties from Unlabeled Videos
- Attribute Embedding with Visual-Semantic Ambiguity Removal for Zero-shot Learning
- NRSfM-Flow: Recovering Non-Rigid Scene Flow from Monocular Image Sequences
- Better Together: Joint Reasoning for Non-rigid 3D Reconstruction with Specularities and Shading
- STAR-Net: A SpaTial Attention Residue Network for Scene Text Recognition
- Context Matters: Refining Object Detection in Video with Recurrent Neural Networks
- Outlier Rejection for Absolute Pose Estimation with Known Orientation
- Learning from scratch a confidence measure
- Localizing Periodicity in Time Series and Videos
- Person Re-identification in Appearance Impaired Scenarios
- Learning to Detect and Match Keypoints with Deep Architectures
- Supervised Incremental Hashing
- Multi-Scale Fully Convolutional Network for Fast Face Detection
- Attention Networks for Weakly Supervised Object Localization
- Image Captioning with Sentiment Terms via Weakly-Supervised Sentiment Dataset
- Learning of Separable Filters by Stacked Fisher Convolutional Autoencoders
- Deskewing by space-variant deblurring
- Faces in Places: compound query retrieval
- Semi-supervised Video Object Segmentation Using Multiple Random Walkers
- Deep Learning for Detecting Multiple Space-Time Action Tubes in Videos
- Exploiting Random RGB and Sparse Features for Camera Pose Estimation
- Fine-grained Recognition in the Noisy Wild: Sensitivity Analysis of Convolutional Neural Networks Approaches
- Near-Field Photometric Stereo in Ambient Light
3D Computer Vision
- Adding Synchronization and Rolling Shutter in Multi-Camera Bundle Adjustment
- EMVS: Event-based Multi-View Stereo
- Occlusion-aware 3D Morphable Face Models
- Next-Best Stereo: Extending Next-Best View Optimisation For Collaborative Sensors
Motion and Tracking
- Shape-based Image Correspondence
- Reprojection Flow for Image Registration Across Seasons
- Unsupervised Learning of Shape-Motion Patterns for Objects in Urban Street Scenes
Keynote 2
Statistical Methods and Learning
- Efficient Learning for Discriminative Segmentation with Supermodular Losses
- Variational Weakly Supervised Gaussian Processes
Recognition and Physics-based vision
- Regional Gating Neural Networks for Multi-label Image Classification
- Multispectral Deep Neural Networks for Pedestrian Detection
- L1 Graph Based Sparse Model for Label De-noising
- PatchIt: Self-Supervised Network Weight Initialization for Fine-grained Recognition
- Combining Shape from Shading and Stereo: A Variational Approach for the Joint Estimation of Depth, Illumination and Albedo
Posters 2
- Solving VIsual Madlibs with Multiple Cues
- LSTM for Image Annotation with Relative Visual Importance
- OnionNet: Sharing Features in Cascaded Deep Classi_ers
- Line reconstruction using prior knowledge in single non-central view
- Attribute Recognition from Adaptive Parts
- Memory-based Gait Recognition
- Three-Point Direct Stereo Visual Odometry
- Semantic Segmentation for Real-World Data by Jointly Exploiting Supervised and Transferrable Knowledge
- Optimized Regressor Forest for Image Super-Resolution
- Convolutional aggregation of local evidence for large pose face alignment
- Wide Residual Networks
- Deep Part-Based Generative Shape Model with Latent Variables
- General Human Traits Oriented Generic Elastic Model for 3D Face Reconstruction
- Attend Refine Repeat: Active Box Proposal Generation via In-Out Localization
- Crafting a multi-task CNN for viewpoint estimation
- Improving Weakly-Supervised Object Localization By Micro-Annotation
- MBest Struct: M-Best diverse sampling for structured tracker
- Event-Based Hough Transform in a Spiking Neural Network for Multiple Line Detection and Tracking Using a Dynamic Vision Sensor
- Holistically Constrained Local Model: Going Beyond Frontal Poses for Facial Landmark Detection
- Bag of Surrogate Parts: one inherent feature of deep CNNs
- Multi-scale Colorectal Tumour Segmentation Using a Novel Coarse to Fine Strategy
- Learning Additive Kernel For Feature Transformation and Its Application to CNN Features
- Robust 3D Car Shape Estimation from Landmarks in Monocular Image
- Projective Unsupervised Flexible Embedding with Optimal Graph
- Towards Automatic Image Editing: Learning to See another You
- Dense Labeling with User Interaction: an Example for Depth-Of-Field Simulation
- MLBoost Revisited: A Faster Metric Learning Algorithm for Identity-Based Face Retrieval
- Learning Neural Network Architectures using Backpropagation
- Enhancing pose estimation through efficient patch synthesis
- Discovering motion hierarchies via tree-structured coding of trajectories
- Coplanar Repeats by Energy Minimization
- Two-Stream SR-CNNs for Action Recognition in Videos
- An Efficient Convolutional Network for Human Pose Estimation
- A data augmentation methodology for training machine/deep learning gait recognition algorithms
- Mean Box Pooling: A Rich Image Representation and Output Embedding for the Visual Madlibs Task
- Practical View on Face Presentation Attack Detection
- Fast Feature-Less Quaternion-based Particle Swarm Optimization for Object Pose Estimation From RGB-D Images
- Graph Convolutional Neural Network
- I Have Seen Enough: Transferring Parts Across Categories
- Impatient DNNs - Deep Neural Networks with Dynamic Time Budgets
- Maximum Margin Linear Classifiers in Unions of Subspaces
- Aerial image geolocalization from recognition and matching of roads and intersections
- Learning local feature descriptors with triplets and shallow convolutional neural networks
- Online Feature Selection for Visual Tracking
- Deep Aggregation of Local 3D Geometric Features for 3D Model Retrieval
- Learning Grimaces by Watching TV
- Accurate and robust face recognition from RGB-D images with a deep learning approach
- Global Deconvolutional Networks for Semantic Segmentation
- Convolutional Sparse Coding-based Image Decomposition
- Domain Adaptive Subspace Clustering
- Filtering 3D Keypoints Using GIST For Accurate Image-Based Localization
- Probabilistic Obstacle Partitioning of Monocular Video for Autonomous Vehicles
- Track Facial Points in Unconstrained Videos
Recognition
- Structured Prediction of 3D Human Pose with Deep Neural Networks
- Multi-task Relative Attribute Prediction by Incorporating Local Context and Global Style Information
- Deep Multi-task Attribute-driven Ranking for Fine-grained Sketch-based Image Retrieval
- The Role of Context Selection in Object Detection
Face and Gesture
- Highly Efficient Regression for Scalable Person Re-Identification
- Reflective Regression of 2D-3D Face Shape Across Large Pose
- Deep Sign: Hybrid CNN-HMM for Continuous Sign Language Recognition
Recognition, Optimisation and Performance Evaluation
- Measuring the effect of nuisance variables on classifiers
- Learning Robust Graph Regularisation for Subspace Clustering
- Solving Jigsaw Puzzles with Linear Programming
- Detecting tracking errors via forecasting
- Oracle Performance for Visual Captioning
Video events, robot vision and deep learning