Seeing things clearly: Image understanding through hard-attention and reasoning with structured knowledges
Computer Vision seminar by Jonathan Gerrand, University of the Witwatersrand
Hosted by SisonkeBiotik Monthly Meets
Thursday 21:00–22:30 Johannesburg (GMT+2)
Ended
Braamfontein, Johannesburg, South Africa · Hybrid
Abstract
In this talk, Jonathan aims to frame the current challenges of explainability and understanding in ML-driven approaches to image processing, and their potential solution through explicit inference techniques.
Topics
africaartificial intelligenceattentioncomputational modelsexplainabilityhard-attentionhealthcareimage processing
Show 7 more topics
inference techniquesmachine learningmedcinemedical imagingreasoningstructured knowledgevisual understanding
Related seminars
SeminarSeminarSeminar
NMC4 Short Talk: Image embeddings informed by natural language improve predictions and understanding of human higher-level visual cortex
Related research
Recording · Dec 1, 2021
Towards a Theory of Human Visual Reasoning
Related research
Recording · Oct 14, 2021
Implementing structure mapping as a prior in deep learning models for abstract reasoning
Related research
Recording · Mar 3, 2022