Bengaluru, India

Sarvesh

This inventor holds 1 USPTO granted patent. Top assignee: Samsung Electronics Ltd.. Active years: 2026.

IDiyas Innovation Intelligence. (2026). Inventor Profile: Sarvesh. Retrieved from https://idiyas.com/inventor/sarvesh-6eo046aq

Data source: USPTO (United States Patent and Trademark Office) public patent records. Weekly synchronization. How IDiyas builds this profile


Average Co-Inventor Count = 1.0


Company Filing History:


Years Active: 2026

Loading Chart...
1 patent (USPTO):Explore Patents

Title: Sarvesh and On-Device Neural Network Inference

Introduction

Sarvesh is an inventor based in Bengaluru, India, whose patent work addresses efficient multi-frame processing in neural networks. He is associated with Samsung Electronics Ltd. and is credited with 1 patent.

Latest Patents

Sarvesh’s patent, “On-device inference method for multi-frame processing in a neural network,” describes a method for optimizing a neural network model that processes multiple input frames. The method selects a predetermined number of frames, identifies a sequence containing a preceding frame and following frames, and uses filters to remove the preceding frame. It then concatenates the following frames in order with the input frames for subsequent processing by the model’s convolution layers.

Career Highlights

The patent reflects work on on-device neural network inference and the organization of frame sequences for multi-frame processing. Sarvesh’s listed company affiliation is Samsung Electronics Ltd.

Collaborations

Sarvesh’s listed coworkers include Kinsuk Das and Raj Narayana Gadde. The provided information does not specify their roles or contributions to the patent.

Conclusion

Sarvesh’s patent focuses on optimizing how neural networks process sequences of video frames, highlighting his contribution to on-device inference technology.

Profile summary based on public USPTO records.
Data Sources: USPTO Patent Grant XML, Patent Center, EPO & CIPO • Normalized by IDiyas Innovation Graph. Methodology & provenance architecturePlease report any incorrect information to [email protected]
Loading…