Samvit Jain

I work at Amazon Web Services (AWS) in Seattle. Here's a broad list of topics I'm currently interested in.

I have previously spent time at Microsoft Research and Databricks. Prior to that, I co-founded LinkMeUp.

I have a B.S.E. in computer science with highest honors from Princeton University, and an M.S. in computer science from UC Berkeley. At Berkeley, I was part of the RISE Lab.

Email / Blog / LinkedIn / GitHub / Twitter
Coaching / Investing
Publications
Accel: A Corrective Fusion Network for Efficient Semantic Segmentation on Video
Samvit Jain, Xin Wang, Joseph Gonzalez
CVPR 2019 [Oral presentation]
summary | arXiv | program | project page | code | video

We present Accel, a novel corrective fusion network that combines (1) optical flow-based feature warping with (2) lightweight, per-frame, temporal correction to achieve state-of-the-art accuracy and throughput on video semantic segmentation.

Spatula: Efficient Cross-Camera Video Analytics on Large Camera Networks
Samvit Jain, Xun Zhang, Yuhao Zhou, Ganesh Ananthanarayanan, Junchen Jiang, Yuanchao Shu, Victor Bahl, Joseph Gonzalez
ACM/IEEE Symposium on Edge Computing (SEC) 2020 [Best paper award]
summary | arXiv | program | website

Enterprises are increasingly deploying large camera networks for video analytics. But video analytics is compute and data intensive, with cost growing with the number of cameras and operation time. To address this cost challenge, we present Spatula, a new system for efficient cross-camera video analytics. Spatula exploits spatial and temporal locality in the dynamics of real camera networks to guide its inference-time search for a query identity.

Scaling Video Analytics Systems to Large Camera Deployments
Samvit Jain, Ganesh Ananthanarayanan, Junchen Jiang, Yuanchao Shu, Joseph Gonzalez
ACM HotMobile 2019
summary | arXiv | program | website | video

We discuss the potential of spatio-temporal correlations -- content correlations between geographically proximate cameras in wide-area enterprise camera deployments -- to improve cost efficiency and inference accuracy in large-scale video analytics operations. Our template application is real-time person re-identification and tracking.

Fast Semantic Segmentation on Video Using Motion Vector-Based Feature Interpolation
Samvit Jain, Joseph Gonzalez
ECCV International Workshop on Video Segmentation 2018
summary | arXiv | program

We exploit video compression techniques (in particular, the block motion vectors in H.264 video) and feature similarity across frames to accelerate a classical scene recognition task, semantic segmentation, on video.

Determining an Optimal Threshold on the Online Reserves of a Bitcoin Exchange
Samvit Jain, Edward Felten, Steven Goldfeder
Journal of Cybersecurity (JCS), 2018
summary | pdf | program

We investigate the fundamental tradeoff between exposure to online (network-based) and offline threats faced by a Bitcoin exchange that must store Bitcoin across online and offline storage, while guaranteeing availability to customers. Parameterizing deposit, withdrawal, and theft events as Poisson processes, we are able to model the financial dynamics of the exchange, and solve for the optimal threshold on online storage.


Technical Reports
Efficient Inference on Video, In Real-Time and At Scale
Samvit Jain
UC Berkeley Master's Thesis (Advisor: Joseph Gonzalez), 2019
summary | pdf

This body of work is motivated by a simple statement: machine learning systems must meet the performance requirements of the applications they enable. Advances in deep learning applied to vision have unlocked opportunity in robotic navigation, industrial and agricultural monitoring, and retail intelligence, each use case with its own latency, throughput, and cost constraints. This thesis is a step toward solving this constrained optimization problem.

Portal: Micropayments on the Paywalled Internet
Samvit Jain
Princeton Senior Thesis (Advisor: Brian Kernighan), 2017
summary | pdf | slides

Portal is a payment protocol and software system that enables one-click purchases of long-form news content on the Internet, without requiring a user to sign up for a subscription or login to a content provider's website.

Monetization on the Modern Web: Automated Micropayments From Bitcoin-Enabled Browsers
Samvit Jain
Princeton Junior Independent Work (Advisor: Arvind Narayanan), 2016
summary | pdf | github

We propose a Bitcoin micropayments-based revenue system for online businesses, which enables users to make small payments to access web content on a per-use basis, in lieu of viewing ads or signing up for a credit card subscription.

Determining an Optimal Threshold on the Online Reserves of a Bitcoin Exchange
Samvit Jain
Princeton Sophomore Independent Work (Advisor: Edward Felten), 2015
summary | pdf | github

We investigate the fundamental tradeoff between exposure to online (network-based) and offline threats faced by a Bitcoin exchange that must store Bitcoin across online and offline storage, while guaranteeing availability to customers. Parameterizing deposit, withdrawal, and theft events as Poisson processes, we are able to model the financial dynamics of the exchange, and solve for the optimal threshold on online storage.

Recent Talks
CVPR 2019 Accel: A Corrective Fusion Network [video] Long Beach, CA June 20, 2019
RISE Lab Retreat Scaling Video Analytics [video] Lake Tahoe, NV May 23, 2019
BAIR Lab Retreat Accel: A Corrective Fusion Network Sonoma, CA Mar 25, 2019
HotMobile 2019 Scaling Video Analytics Santa Cruz, CA Feb 27, 2019
ECCV 2018 Workshop Accelerating Semantic Segmentation on Video Munich, Germany Sep 14, 2018