Description

This course studies learning visual representations for common computer vision tasks including matching, retrieval, classification, and object detection. Related problems are discussed including indexing, nearest neighbor search, clustering, and dimensionality reduction. The course discusses well-known methods from low-level description to intermediate representation, and their dependence on the end task. It then studies a data-driven approach where the entire pipeline is optimized jointly in a supervised fashion, according to a task-dependent objective. Deep learning models are studied in detail and interpreted in connection to conventional models. The focus of the course is on recent, state of the art methods and large scale applications.

The course is part of master program Research in Computer Science (SIF) of University of Rennes 1.

The following refers to the first iteration of the course in Nov. 2017 - Jan. 2018.

Instructor

Yannis Avrithis

Discussions

Piazza

Class

Monday and Wednesday
16:15 - 18:15

Evaluation

Oral presentations: 50%
Written exam: 50%

Planning and Syllabus

Event Date Room Description Material*
Lecture 1 Monday
Nov 20
B02B-E110 (23) Introduction
Research field. Neuroscience, computer vision and machine learning background. Modern deep learning. About this course.
[slides]
Lecture 2 Wednesday
Nov 22
B12D i-59 (44) Representation
Global/local visual descriptors, dense/sparse representation, feature detectors. Encoding/pooling, vocabularies, bag-of-words. Match kernels, embedding, Fisher vectors, VLAD.
[slides]
Lecture 3 Monday
Nov 27
B12D i-58 (44) Matching
Spatial matching, geometric models, RANSAC, Hough transform. Pyramid matching, spatial and Hough pyramids. Object detection, subwindow search, Hough model, deformable part model.
[slides]
Lecture 4 Wednesday
Dec 6
B12D i-58 (44) Indexing
Clustering, dimensionality reduction, density estimation, nearest neighbor search. Tree-based methods, hashing, product quantization. Inverted index and multi-index.
[slides]
Lecture 5 Monday
Dec 11
B12D i-59 (44) Learning
Naive Bayes, nearest neighbor classification. Regression, classification. Logistic regression, support vector machines, neural networks. Activation functions, loss functions, gradient descent.
[slides]
Lecture 6 Wednesday
Dec 13
B12D i-58 (44) Differentiation
Computational graphs, back-propagation, automatic differentiation.
[slides]
Lecture 7 Monday
Dec 18
B12D i-58 (44) Convolution
Pooling, strided convolution, dilated convolution. Convolutional networks. Deconvolution, fully convolutional networks.
[slides]
Milestone Tuesday
Dec 19
Oral presentation
Selection of papers to present.
[instructions]
Lecture 8 Wednesday
Jan 10
B12D i-58 (44) Optimization
Parameter initialization, data-dependent initialization, normalization, regularization. Optimization methods, second-order methods, Hessian-free methods.
[slides]
Lecture 9 Monday
Jan 15
B12D i-52 (44) Detection
Class-agnostic region proposals, bounding box regression, non-maxima suppression, part-based models, spatial transformers, attention networks.
[slides]
Lecture 10 Monday
Jan 22
B12D i-52 (44) Retrieval
Siamese, triplet, and batch-wise loss functions. Embedding, pooling, dimensionality reduction and manifold learning. Partial matching, spatial matching, quantization, diffusion.
[slides]
Evaluation 1 Wednesday
Jan 24
B12D i-58 (44) Written exam
Evaluation 2 Monday
Jan 29
Jersey Oral presentations
08:00 - 12:00
[instructions]

*All material licensed CC BY-SA 4.0

Prerequisites


Basic knowledge of Linear Algebra, Calculus, Probabilities, Machine Learning, Signal Processing, Python.