Discussing the article: "Neural Networks Made Easy (Part 81): Context-Guided Motion Analysis (CCMR)"
You are missing trading opportunities:
- Free trading apps
- Over 8,000 signals for copying
- Economic news for exploring financial markets
Registration
Log in
You agree to website policy and terms of use
If you do not have an account, please register
Check out the new article: Neural Networks Made Easy (Part 81): Context-Guided Motion Analysis (CCMR).
In previous works, we always assessed the current state of the environment. At the same time, the dynamics of changes in indicators always remained "behind the scenes". In this article I want to introduce you to an algorithm that allows you to evaluate the direct change in data between 2 successive environmental states.
As part of this series, we became acquainted with various methods for analyzing the state of the environment and algorithms for using the obtained data. We used convolutional models to find stable patterns in historical price movement data. We also used attention models to find dependencies between distinct local environmental states. We always assessed the state of the environment as a certain cross-section at a point in time. However, we have never assessed the dynamics of environmental indicators. We assumed that the model, in the process of analyzing and comparing environmental conditions, would somehow pay attention to key changes. But we did not use an explicit quantitative representation of such dynamics.
However, in the field of computer vision, there is a fundamental problem of optical flow estimation. The solution to this problem provides information about the movement of objects in the scene. To solve this problem, a number of interesting algorithms have been proposed and are now widely used. Optical flow estimation results are used in various fields from autonomous driving to object tracking and surveillance.
Most current approaches use convolutional neural networks, but they lack global context. This makes it difficult to reason about object occlusions or large displacements. An alternative approach is to use transformers and other attention techniques. They allow you to go far beyond the fixed receptive field of classical CNNs.
A particularly interesting method entitled CCMR was presented in the paper "CCMR: High Resolution Optical Flow Estimation via Coarse-to-Fine Context-Guided Motion Reasoning". It is an approach to optical flow estimation that combines the advantages of attention-oriented methods of motion aggregation concepts and high-resolution multi-scale approaches. The CCMR method consistently integrates context-based motion grouping concepts into a high-resolution coarse-grained estimation framework. This allows for detailed flow fields that also provide high accuracy in occluded areas. In this context, the authors of the method propose a two-stage motion grouping strategy where global self-attentional contextual features are first computed and them used to guide motion features iteratively across all scales. Thus, context-directed reasoning about XCiT-based motion provides processing at all coarse-grained scales. Experiments conducted by the authors of the method demonstrate the strong performance of the proposed approach and the advantages of its basic concepts.
Author: Dmitriy Gizlyk