Data augmentation is the process of transforming training data to introduce randomness. This strategy is common for computer vision tasks. In this scenario, the training data in question are images. For example, you can scale, rotate, mirror, and/or crop your images during training. Image augmentation has two key benefits: One…

Object detection is the process of analyzing images or videos with the purpose of identifying the various classes of objects present and a bounding box to identify their position. This concept has been quite popular in the machine learning community, with the mainstream solutions offering 2D bounding boxes. …

Voice recognition software allows people to tell their devices what to do by just talking to them, without the need to use a mouse, keyboard, or touch a screen. This kind of interaction with computers is dramatically changing the way we experience technology by lowering adoption barriers, helping with multitasking…

Facial expressions play an important role in non-verbal communications, especially in conveying emotions like happiness, surprise, shock, sadness, indifference, etc. Selfie effects have been used to visually amplify these expressions. MediaPose has two deep learning models targeting the face: a face detector and a face mesh solution. These two neural…

One of the biggest blowups that came with the smartphone era is the ability to capture selfies, such a powerful concept that Oxford dictionary added the word to the dictionary. This phenomenon led to the growth of social media apps like Snapchat, which innovated the creation of selfies by using…

Hand tracking has proven to be useful in an array of applications like using gestures as input controls, especially in virtual, augmented and mixed reality experiences. This capability is powered by advanced machine learning algorithms that perceive the shape of hands and track the motion of joints. …

Animation offers a unique opportunity for storytelling, but it is a time-consuming and expensive task that requires an individual to have specialized skills. This has made creating animated character videos only accessible to the top-tier brands who can afford to hire an animation studio. To animate a character, you must…

Mobile apps nowadays are shipped with machine intelligence that let the user achieve more with less friction. Some of these intelligent features come in the form of object recognition, speech analysis, image segmentation, predictions and so much more. There are two ways of deploying such intelligence: cloud-based or on-device. …

The machine learning community is one of the most progressive communities in the world as people and organizations involved are constantly improving skills, research, and machine learning models. This has led to the publication of numerous papers, open-source tools, and models that make the lives of ML engineers easier. The…