Texas Instruments Corporation

Shyam Jagannathan

Pramod Kumar Swami

Naveen Srinivasamurthy

Mihir Narendra Mody

Soyeb Noormohammed Nagori

Manu Mathew

Deepak Kumar Poddar

Kedar Satish Chitnis

Timothy David Anderson

Arun Shankar Kudana

Son Hung Tran

Manoj Koul

Jason T Jones

Niraj Nandan

Hrushikesh Tukaram Garud

Kumar Arrakutti Desappan

Prithvi Shankar Yeyyadi Anantha

Brijesh Rameshbhai Jadav

David E Smith

Kumar Desappan

Ankur Ankur

Soyeb Noomohammed Nagori

Vivek Vilas Dhande

Santhanakrishnan Narayanan Narayanan

David Smith

Techniques for executing machine learning (ML) models including receiving an indication to run an ML model on a processing core; receiving a static memory allocation for running the ML model on the processing core; determining that a layer of the ML model uses more memory than the static memory allocated; transmitting, to a shared memory, a memory request for blocks of the shared memory; receiving an allocation of the requested blocks; running the layer of the ML model using the static memory and the range of memory addresses; and outputting results of running the layer of the ML model.

Memory sharing for machine learning processing

Described examples include an integrated circuit including a vector multiply unit including a plurality of multiply/accumulate nodes, in which the vector multiply unit is operable to provide an output from the multiply/accumulate nodes, a first data feeder operable to provide first data to the vector multiply unit in vector format, and a second data feeder operable to provide second data to the vector multiply unit in vector format.

Methods and apparatus for matrix processing in a convolutional neural network

A video encoder including a first buffer containing a plurality of data values defining a macroblock of pixels of a video frame. The video encoder also includes a second buffer and an entropy encoder coupled to the first and second buffers and configured to encode a macroblock based on another macroblock. The entropy encoder identifies a subset of the data values from the first buffer defining a given macroblock and copies the identified subset to the second buffer, the subset of data values being just those data values used by the entropy encoder when subsequently encoding another macroblock.

Saving minimum macroblock data for subsequent encoding of other macroblocks

Software instructions are executed on a processor within a computer system to configure a streaming engine with stream parameters to define a multidimensional array. The stream parameters define a size for each dimension of the multidimensional array and a specified width for a selected dimension of the array. Data is fetched from a memory coupled to the streaming engine responsive to the stream parameters. A stream of vectors is formed for the multidimensional array responsive to the stream parameters from the data fetched from memory. When the selected dimension in the stream of vectors exceeds the specified width, the streaming engine inserts null elements into each portion of a respective vector for the selected dimension that exceeds the specified width in the stream of vectors. Stream vectors that are completely null are formed by the streaming engine without accessing the system memory for respective data.

One-dimensional zero padding in a stream of matrix elements

A system-on-chip (SoC) in which trace data is managed includes a first memory device, a first interface to couple the first memory to a second memory external to the system-on-chip, and a first processing resource coupled to the first interface and the first memory device. The first processing resource includes a data buffer and a first direct access memory (DMA) controller. The first DMA controller transmits data from the data buffer to the first interface over a first channel, and transmits the data from the data buffer with associated trace information for the data to the first memory device over a second channel.

Multi-channel system architecture for managing flow of data and associated trace information

The disclosure provides a noise filter. The noise filter includes a motion estimation (ME) engine. The ME receives a current frame and a reference frame. The current frame comprising a current block and the reference frame includes a plurality of reference blocks. The ME engine generates final motion vectors. The current block comprises a plurality of current pixels. A motion compensation unit generates a motion compensated block based on the final motion vectors and the reference frame. The motion compensated block includes a plurality of motion compensated pixels. A weighted average filter multiplies each current pixel of the plurality of current pixels and a corresponding motion compensated pixel of the plurality of motion compensated pixels with a first weight and a second weight respectively. The weighted average filter generates a filtered block. A blockiness removal unit is coupled to the weighted average filter and removes artifacts in the filtered block.

Method and apparatus for noise reduction in video systems

Disclosed examples include image processing methods and systems to process image data, including computing a plurality of scaled images according to input image data for a current image frame, computing feature vectors for locations of the individual scaled images, classifying the feature vectors to determine sets of detection windows, and grouping detection windows to identify objects in the current frame, where the grouping includes determining first clusters of the detection windows using non-maxima suppression grouping processing, determining positions and scores of second clusters using mean shift clustering according to the first clusters, and determining final clusters representing identified objects in the current image frame using non-maxima suppression grouping of the second clusters. Disclosed examples also include methods and systems to track identified objects from one frame to another using feature vectors and overlap of identified objects between frames to minimize computation intensive operations involving feature vectors.

Window grouping and tracking for fast object detection

Software instructions are executed on a processor within a computer system to configure a steaming engine with stream parameters to define a multidimensional array. The stream parameters define a size for each dimension of the multidimensional array and a specified width for a selected dimension of the array. Data is fetched from a memory coupled to the streaming engine responsive to the stream parameters. A stream of vectors is formed for the multidimensional array responsive to the stream parameters from the data fetched from memory. When the selected dimension in the stream of vectors exceeds the specified width, the streaming engine inserts null elements into each portion of a respective vector for the selected dimension that exceeds the specified width in the stream of vectors. Stream vectors that are completely null are formed by the streaming engine without accessing the system memory for respective data.

A method for object classification in a decision tree based adaptive boosting (AdaBoost) classifier implemented on a single-instruction multiple-data (SIMD) processor is provided that includes receiving feature vectors extracted from N consecutive window positions in an image in a memory coupled to the SIMD processor and evaluating the N consecutive window positions concurrently by the AdaBoost classifier using the feature vectors and vector instructions of the SIMD processor, in which the AdaBoost classifier concurrently traverses decision trees for the N consecutive window positions until classification is complete for the N consecutive window positions.

Growing community of inventors

Bengaluru, India

Shyam Jagannathan