A Complete Guide for Camera Calibration Guide in 2024
Viso.ai
by Nico Klingler
3d ago
One way to think of a camera is a device that projects a 3D world in a 2D space. Humans are capable of looking at these 2D images and intuitively inferring the 3D world. For example, the relative distances, size movements, and spatial relationships of objects in the images. However, computers do not have this inherent ability. What is Camera Calibration? Camera calibration is the process of identifying the geometric characteristics of 2D images captured in 3D space. This allows image processing systems to make inferences about the scenes in these images for applications where metric informatio ..read more
Visit website
What are Liquid Neural Networks?
Viso.ai
by Gaudenz Boesch
3d ago
Neural Networks have changed the way we perform model training. This gave birth to a new domain called Deep Learning. With the increase in the usage of the internet and computer systems, the availability of data has become easy. Neural networks, sometimes referred to as Neural Nets, need large datasets for efficient training. This availability of data made it easy to train and adopt across the industry. Even though they work efficiently, they do have some challenges. For example, if the model we trained encounters something outside the training set, it won’t work properly as it has never encou ..read more
Visit website
Xception Model: A Deep Dive into Depthwise Separable Convolutions
Viso.ai
by Gaudenz Boesch
3d ago
Xception, short for Extreme Inception, is a Deep Learning model that is developed by Francois Chollet at Google, continuing the popularity of Inception architecture, and further perfecting it. The inception architecture utilizes inception modules, however, the Xception model replaces it with depthwise separable convolution layers, which totals 36 layers. When we compare the Xception model with the Inception V3 model, it only slightly performs better on the ImageNet dataset, however, on larger datasets consisting of 350 million images, Xception performs significantly better. The journey of Deep ..read more
Visit website
Image Fusion in Computer Vision
Viso.ai
by Nico Klingler
1w ago
In many computer vision applications (e.g. robot motion and medical imaging) there is a need to integrate relevant information from multiple images into a single image. Such image fusion will provide higher reliability, accuracy, and data quality. Multiview fusion improves the image with higher resolution and also recovers the 3D representation of a scene. Multimodal fusion combines images from different sensors and is referred to as multi-sensor fusion. Its main applications include medical imagery, surveillance and security. About us: Viso.ai provides a robust end-to-end no-code computer vis ..read more
Visit website
Text Annotation: The Complete Guide
Viso.ai
by Nico Klingler
1w ago
Data annotation allows machine learning algorithms to understand and interpret information. The annotations are labels that identify and classify data or associate different pieces of information with each other. AI algorithms use them as ground truths to adjust their weights accordingly. The labels are task-dependent and can be further categorized as an image or text annotation. Text annotations associate meaning with textual information for ML algorithms to understand. They generate labels that allow ML algorithms to interpret the text in a human-like fashion. The process involves classifyin ..read more
Visit website
Typical Workflow for Building a Machine Learning Model
Viso.ai
by Nico Klingler
1w ago
Whether you’re building a consumer app to recognize plant species or an enterprise tool to monitor office security camera footage, you are going to need to build a Machine Learning (ML) model to provide the core functionality.  Today, building an ML model is easier than ever before using frameworks like Tensorflow.  However, it’s still important to follow a methodical workflow to avoid building a model with poor performance or inherent bias. Building a machine learning model consists of 7 high-level steps: 1. Problem Identification 2. Dataset Creation 3. Model Selection 4. Model Trai ..read more
Visit website
Neural Radiance Fields (NeRFs): A Technical Exploration
Viso.ai
by Gaudenz Boesch
1w ago
Neural Radiance Fields (NeRFs) are a deep learning technique that uses fully connected neural networks to represent 3D scenes from a collection of 2D images and then render them to synthesize novel views. With their ability to synthesize realistic new viewpoints, Neural Radiance Fields excel in a wide range of 3D capture scenarios, from panoramic views of enclosed spaces to vast, open landscapes. Recall your last vacation where you captured a few photos of your favorite place. NeRFs offer the potential to revisit such scenes in immersive 3D, generating realistic views from multiple angles. Thi ..read more
Visit website
Decoding Movement: Spatio-Temporal Action Recognition
Viso.ai
by Gaudenz Boesch
2w ago
Introduction to Spatio-Temporal Action Recognition Fundamentals Many use the terms Spatio-Temporal Action Recognition, localization, and detection interchangeably. However, there is a subtle difference in exactly what they focus on. Spatio-temporal action recognition identifies both the type of action that occurs and when it happens. Localization comprises the recognition as well as pinpointing its spatial location within each frame over time. Detection focuses on when an action starts and ends and how long it lasts in a video. Let’s take an example of a video clip featuring a running man. Rec ..read more
Visit website
MobileNet – Efficient Deep Learning for Mobile Vision
Viso.ai
by Nico Klingler
2w ago
MobileNet was developed by a team of researchers at Google in 2017, who aimed to design an efficient Convolution Neural Network (CNN) for mobile and embedded devices. The model they created was not only significantly smaller in size and efficient but was also at par with top models in terms of performance. Today, MobileNet is used in various real-world applications to perform object detection and image classification in facial recognition, augmented reality, and more. In this blog, we will look into how MobileNet was able to bring down the total number of parameters by almost 10 times, and als ..read more
Visit website
Home Robots: the Stanford’s Roadmap Paper
Viso.ai
by Nico Klingler
3w ago
Stanford University and panel researchers P. Stone and R. Brooks et al. (2016) created the report “One Hundred Year Study on Artificial Intelligence (AI100)”. The study panel summarized the current progress and envisioned future advancements in the areas of AI and home robots as follows: AI and computer vision’s advancement led the gaming industry to surpass Hollywood as the biggest entertainment industry. Deep learning and Convolutional Neural Networks (CNNs) have enabled speech understanding and computer vision on our phones, cars, and homes. Natural Language Processing (NLP) and knowledge ..read more
Visit website

Follow Viso.ai on FeedSpot

Continue with Google
Continue with Apple
OR