Deep learningDeep learning is part of a broader family of machine learning methods, which is based on artificial neural networks with representation learning. The adjective "deep" in deep learning refers to the use of multiple layers in the network. Methods used can be either supervised, semi-supervised or unsupervised.
CameraA camera is an optical instrument used to capture and store images or videos, either digitally via an electronic , or chemically via a light-sensitive material such as photographic film. As a pivotal technology in the fields of photography and videography, cameras have played a significant role in the progression of visual arts, media, entertainment, surveillance, and scientific research. The invention of the camera dates back to the 19th century and has since evolved with advancements in technology, leading to a vast array of types and models in the 21st century.
Feedforward neural networkA feedforward neural network (FNN) is one of the two broad types of artificial neural network, characterized by direction of the flow of information between its layers. Its flow is uni-directional, meaning that the information in the model flows in only one direction—forward—from the input nodes, through the hidden nodes (if any) and to the output nodes, without any cycles or loops, in contrast to recurrent neural networks, which have a bi-directional flow.
Convolutional neural networkConvolutional neural network (CNN) is a regularized type of feed-forward neural network that learns feature engineering by itself via filters (or kernel) optimization. Vanishing gradients and exploding gradients, seen during backpropagation in earlier neural networks, are prevented by using regularized weights over fewer connections. For example, for each neuron in the fully-connected layer 10,000 weights would be required for processing an image sized 100 × 100 pixels.
Artificial neural networkArtificial neural networks (ANNs, also shortened to neural networks (NNs) or neural nets) are a branch of machine learning models that are built using principles of neuronal organization discovered by connectionism in the biological neural networks constituting animal brains. An ANN is based on a collection of connected units or nodes called artificial neurons, which loosely model the neurons in a biological brain. Each connection, like the synapses in a biological brain, can transmit a signal to other neurons.
3D scanning3D scanner is the process of analyzing a real-world object or environment to collect three dimensional data of its shape and possibly its appearance (e.g. color). The collected data can then be used to construct digital 3D models. A 3D scanner can be based on many different technologies, each with its own limitations, advantages and costs. Many limitations in the kind of objects that can be digitised are still present. For example, optical technology may encounter many difficulties with dark, shiny, reflective or transparent objects.
Types of artificial neural networksThere are many types of artificial neural networks (ANN). Artificial neural networks are computational models inspired by biological neural networks, and are used to approximate functions that are generally unknown. Particularly, they are inspired by the behaviour of neurons and the electrical signals they convey between input (such as from the eyes or nerve endings in the hand), processing, and output from the brain (such as reacting to light, touch, or heat). The way neurons semantically communicate is an area of ongoing research.
Time-of-flight cameraA time-of-flight camera (ToF camera), also known as time-of-flight sensor (ToF sensor), is a range imaging camera system for measuring distances between the camera and the subject for each point of the image based on time-of-flight, the round trip time of an artificial light signal, as provided by a laser or an LED. Laser-based time-of-flight cameras are part of a broader class of scannerless LIDAR, in which the entire scene is captured with each laser pulse, as opposed to point-by-point with a laser beam such as in scanning LIDAR systems.
Pose (computer vision)In the fields of computing and computer vision, pose (or spatial pose) represents the position and orientation of an object, usually in three dimensions. Poses are often stored internally as transformation matrices. The term “pose” is largely synonymous with the term “transform”, but a transform may often include scale, whereas pose does not. In computer vision, the pose of an object is often estimated from camera input by the process of pose estimation.
Deep belief networkIn machine learning, a deep belief network (DBN) is a generative graphical model, or alternatively a class of deep neural network, composed of multiple layers of latent variables ("hidden units"), with connections between the layers but not between units within each layer. When trained on a set of examples without supervision, a DBN can learn to probabilistically reconstruct its inputs. The layers then act as feature detectors. After this learning step, a DBN can be further trained with supervision to perform classification.
Single-lens reflex cameraA single-lens reflex camera (SLR) is a camera that typically uses a mirror and prism system (hence "reflex" from the mirror's reflection) that permits the photographer to view through the lens and see exactly what will be captured. With twin lens reflex and rangefinder cameras, the viewed image could be significantly different from the final image. When the shutter button is pressed on most SLRs, the mirror flips out of the light path, allowing light to pass through to the light receptor and the image to be captured.
Digital single-lens reflex cameraA digital single-lens reflex camera (digital SLR or DSLR) is a digital camera that combines the optics and the mechanisms of a single-lens reflex camera with a solid-state and digitally records the images from the sensor. The reflex design scheme is the primary difference between a DSLR and other digital cameras. In the reflex design, light travels through the lens and then to a mirror that alternates to send the image to either a prism, which shows the image in the optical viewfinder, or the image sensor when the shutter release button is pressed.
Pose trackingIn virtual reality (VR) and augmented reality (AR), a pose tracking system detects the precise pose of head-mounted displays, controllers, other objects or body parts within Euclidean space. Pose tracking is often referred to as 6DOF tracking, for the six degrees of freedom in which the pose is often tracked. Pose tracking is sometimes referred to as positional tracking, but the two are separate. Pose tracking is different from positional tracking because pose tracking includes orientation whereas and positional tracking does not.
Deep reinforcement learningDeep reinforcement learning (deep RL) is a subfield of machine learning that combines reinforcement learning (RL) and deep learning. RL considers the problem of a computational agent learning to make decisions by trial and error. Deep RL incorporates deep learning into the solution, allowing agents to make decisions from unstructured input data without manual engineering of the state space. Deep RL algorithms are able to take in very large inputs (e.g.
Recurrent neural networkA recurrent neural network (RNN) is one of the two broad types of artificial neural network, characterized by direction of the flow of information between its layers. In contrast to uni-directional feedforward neural network, it is a bi-directional artificial neural network, meaning that it allows the output from some nodes to affect subsequent input to the same nodes. Their ability to use internal state (memory) to process arbitrary sequences of inputs makes them applicable to tasks such as unsegmented, connected handwriting recognition or speech recognition.
Camera resectioningCamera resectioning is the process of estimating the parameters of a pinhole camera model approximating the camera that produced a given photograph or video; it determines which incoming light ray is associated with each pixel on the resulting image. Basically, the process determines the pose of the pinhole camera. Usually, the camera parameters are represented in a 3 × 4 projection matrix called the camera matrix. The extrinsic parameters define the camera pose (position and orientation) while the intrinsic parameters specify the camera image format (focal length, pixel size, and image origin).
Light field cameraA light field camera, also known as a plenoptic camera, is a camera that captures information about the light field emanating from a scene; that is, the intensity of light in a scene, and also the precise direction that the light rays are traveling in space. This contrasts with conventional cameras, which record only light intensity at various wavelengths. One type uses an array of micro-lenses placed in front of an otherwise conventional image sensor to sense intensity, color, and directional information.
Dimension (data warehouse)A dimension is a structure that categorizes facts and measures in order to enable users to answer business questions. Commonly used dimensions are people, products, place and time. (Note: People and time sometimes are not modeled as dimensions.) In a data warehouse, dimensions provide structured labeling information to otherwise unordered numeric measures. The dimension is a data set composed of individual, non-overlapping data elements. The primary functions of dimensions are threefold: to provide filtering, grouping and labelling.
Mirrorless cameraA mirrorless camera is a camera that does not have a mirror. Though most cameras, including smartphones, lack mirrors, the term mirrorless is primarily used to describe digital interchangeable lens cameras, in order to distinguish them from DSLRs, which have historically dominated interchangeable lens cameras. Lacking a mirror system allows the camera to be smaller, quieter, and lighter. In cameras with mirrors, light from the lens is directed to either the image sensor or the viewfinder.
Residual neural networkA Residual Neural Network (a.k.a. Residual Network, ResNet) is a deep learning model in which the weight layers learn residual functions with reference to the layer inputs. A Residual Network is a network with skip connections that perform identity mappings, merged with the layer outputs by addition. It behaves like a Highway Network whose gates are opened through strongly positive bias weights. This enables deep learning models with tens or hundreds of layers to train easily and approach better accuracy when going deeper.
