Abstract: System and method for extracting human pose information from an image, comprising a feature extractor connected to a database, a convolutional neural network (CNN) with a plurality of CNN layers. Said system/method further comprising at least one of the following modules: a 2D body skeleton detector for determining 2D body skeleton information from the human-related image features; a body silhouette detector for determining body silhouette information from the human-related image features; a hand silhouette detector for determining hand silhouette detector from the human-related image features; a hand skeleton detector for determining hand skeleton from the human-related image features; a 3D body skeleton detector for determining 3D body skeleton from the human-related image features; and a facial keypoints detector for determining facial keypoints from the human-related image features.

Abstract: An activity classifier system and method that classifies human activities using 2D skeleton data. The system includes a skeleton preprocessor that transforms the 2D skeleton data into transformed skeleton data, the transformed skeleton data comprising scaled, relative joint positions and relative joint velocities. The system also includes a gesture classifier comprising a first recurrent neural network that receives the transformed skeleton data, and is trained to identify the most probable of a plurality of gestures. The system also has an action classifier comprising a second recurrent neural network that receives information from the first recurrent neural networks and is trained to identify the most probable of a plurality of actions.

Abstract: An activity classifier system and method that classifies human activities using 2D skeleton data. The system includes a skeleton preprocessor that transforms the 2D skeleton data into transformed skeleton data, the transformed skeleton data comprising scaled, relative joint positions and relative joint velocities. The system also includes a gesture classifier comprising a first recurrent neural network that receives the transformed skeleton data, and is trained to identify the most probable of a plurality of gestures. The system also has an action classifier comprising a second recurrent neural network that receives information from the first recurrent neural networks and is trained to identify the most probable of a plurality of actions.

Abstract: An activity classifier system and method that classifies human activities using 2D skeleton data. The system includes a skeleton preprocessor that transforms the 2D skeleton data into transformed skeleton data, the transformed skeleton data comprising scaled, relative joint positions and relative joint velocities. The system also includes a gesture classifier comprising a first recurrent neural network that receives the transformed skeleton data, and is trained to identify the most probable of a plurality of gestures. The system also has an action classifier comprising a second recurrent neural network that receives information from the first recurrent neural networks and is trained to identify the most probable of a plurality of actions.

Abstract: System and method for extracting human pose information from an image, comprising a feature extractor connected to a database, a convolutional neural network (CNN) with a plurality of CNN layers. Said system/method further comprising at least one of the following modules: a 2D body skeleton detector for determining 2D body skeleton information from the human-related image features; a body silhouette detector for determining body silhouette information from the human-related image features; a hand silhouette detector for determining hand silhouette detector from the human-related image features; a hand skeleton detector for determining hand skeleton from the human-related image features; a 3D body skeleton detector for determining 3D body skeleton from the human-related image features; and a facial keypoints detector for determining facial keypoints from the human-related image features.

Abstract: An activity classifier system and method that classifies human activities using 2D skeleton data. The system includes a skeleton preprocessor that transforms the 2D skeleton data into transformed skeleton data, the transformed skeleton data comprising scaled, relative joint positions and relative joint velocities. The system also includes a gesture classifier comprising a first recurrent neural network that receives the transformed skeleton data, and is trained to identify the most probable of a plurality of gestures. The system also has an action classifier comprising a second recurrent neural network that receives information from the first recurrent neural networks and is trained to identify the most probable of a plurality of actions.

Abstract: This disclosure is directed to an activity classifier system, for classifying human activities using 2D skeleton data. The system includes a skeleton preprocessor that transforms the 2D skeleton data into transformed skeleton data, the transformed skeleton data comprising scaled, relative joint positions and relative joint velocities. It also includes a gesture classifier comprising a first recurrent neural network that receives the transformed skeleton data, and is trained to identify the most probable of a plurality of gestures. There is also an action classifier comprising a second recurrent neural network that receives information from the first recurrent neural networks and is trained to identify the most probable of a plurality of actions.

Abstract: A method and apparatus for calibrating a pipelined analog-to-digital converter (ADC) is disclosed. A method includes reading a first output level from a first sub-ADC, reading one or more additional output levels from one or more additional sub-ADCs, combining the one or more additional output levels from the one or more additional sub-ADCs into a combined output level, and adjusting a comparator threshold of the first sub-ADC when the first output level and the combined output level meet a set of predetermined conditions.

Abstract: An electronic device is disclosed that includes an analog-to-digital converter circuit, an adaptive filter circuit coupled to the analog-to-digital converter circuit to correct one or more circuit impairments in the analog-to-digital converter circuit, a training signal generator circuit to generate training signals, and an amplitude detector circuit configured to suspend generation of the training signals and cause the adaptive filter circuit to suspend adaptation when the input signal is above a predetermined threshold.

Abstract: A method and apparatus for calibrating a CDAC-based analog-to-digital converter is disclosed. In one aspect, a calibration method includes: applying a predetermined pattern of voltages to first plates of a group of N capacitors, wherein N is an integer greater than 1; applying a zero voltage to the second plates of the group of N capacitors, wherein the second plates of the group of N capacitors are connected in common; removing the zero voltage to the second plates of the group of N capacitors; applying a zero voltage to all of the first plates of the group of N capacitors; quantizing a voltage on the second plates of the group of N capacitors; converting the quantized voltage on the second plates of the group of N capacitors to an adjustment value; and loading the adjustment value into a lookup table.

Abstract: A video decoder includes an entropy decoding device that includes a first processor that generates entropy decoded (EDC) data from an encoded video signal that includes a plurality of video layers. A general video decoding device includes a second processor that generates a decoded video signal from the EDC data, wherein the general video decoding device includes a neighbor management module, a decode motion compensation module, an inverse intra-prediction module, an inverse transform/quantization module, a deblocking filter module, and a resampling module.

Abstract: A video decoder includes an entropy decoding device that includes a first processor that generates entropy decoded (EDC) data from an encoded video signal. A general video decoding device includes a second processor that generates a decoded video signal from the EDC data, wherein the general video decoding device includes: a neighbor management module, a decode motion compensation module, an inverse intra-prediction module, an inverse transform/quantization module, and a deblocking filter module.

Abstract: A video decoder includes an entropy decoding device that includes a first processor that generates entropy decoded (EDC) data from an encoded video signal. A general video decoding device includes a second processor that generates a decoded video signal from the EDC data, wherein the general video decoding device includes: a neighbor management module, a decode motion compensation module, an inverse intra-prediction module, an inverse transform/quantization module, and a deblocking filter module.

Abstract: A video decoder includes an entropy decoding device that includes a first processor that generates entropy decoded (EDC) data from an encoded video signal that includes a plurality of video layers. A general video decoding device includes a second processor that generates a decoded video signal from the EDC data, wherein the general video decoding device includes a neighbor management module, a decode motion compensation module, an inverse intra-prediction module, an inverse transform/quantization module, a deblocking filter module, and a resampling module.