Hariprasad Kannan - Academia.edu (original) (raw)

Uploads

Papers by Hariprasad Kannan

Les relaxations en probleme d’optimisation lineaire jouent un role central en inference du maximu... more Les relaxations en probleme d’optimisation lineaire jouent un role central en inference du maximum a posteriori (map) dans les champs aleatoires de Markov discrets. Nous etudions ici les avantages offerts par les methodes de Newton pour resoudre efficacement le probleme dual (au sens de Lagrange) d’une reformulation lisse du probleme. Nous comparons ces dernieres aux methodes de premier ordre, a la fois en terme de vitesse de convergence et de robustesse au mauvais conditionnement du probleme. Nous exposons donc un cadre general pour l’apprentissage non-supervise base sur le transport optimal et les regularisations parcimonieuses. Nous exhibons notamment une approche prometteuse pour resoudre le probleme de la preimage dans l’acp a noyau. Du point de vue de l’optimisation, nous decrivons le calcul du gradient d’une version lisse de la norme p de Schatten et comment cette derniere peut etre utilisee dans un schema de majoration-minimisation.

Solving optimization tasks based on functions and losses with a topological flavor is a very acti... more Solving optimization tasks based on functions and losses with a topological flavor is a very active, growing field of research in data science and Topological Data Analysis, with applications in non-convex optimization, statistics and machine learning. However, the approaches proposed in the literature are usually anchored to a specific application and/or topological construction, and do not come with theoretical guarantees. To address this issue, we study the differentiability of a general map associated with the most common topological construction, that is, the persistence map. Building on real analytic geometry arguments, we propose a general framework that allows us to define and compute gradients for persistence-based functions in a very simple way. We also provide a simple, explicit and sufficient condition for convergence of stochastic subgradient methods for such functions. This result encompasses all the constructions and applications of topological optimization in the lit...

Solving optimization tasks based on functions and losses with a topological flavor is a very acti... more Solving optimization tasks based on functions and losses with a topological flavor is a very active and growing field of research in Topological Data Analysis, with plenty of applications in non-convex optimization, statistics and machine learning. All of these methods rely on the fact that most of the topological constructions are actually stratifiable and differentiable almost everywhere. However, the corresponding gradient and associated code is always anchored to a specific application and/or topological construction, and do not come with theoretical guarantees. In this article, we study the differentiability of a general functional associated with the most common topological construction, that is, the persistence map, and we prove a convergence result of stochastic subgradient descent for such a functional. This result encompasses all the constructions and applications for topological optimization in the literature, and comes with code that is easy to handle and mix with other ...

2006 American Control Conference, 2006

2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Jul 1, 2017

2012 Ninth Conference on Computer and Robot Vision, 2012

We present an algorithm to estimate the pose of a human head from a single image. This algorithm ... more We present an algorithm to estimate the pose of a human head from a single image. This algorithm covers a wide range of head orientations and can be used at various image resolutions. It does not need personalized initialization and is also relatively insensitive to illumination. It works on cropped images obtained by simple head alignment and is fast enough

Handbook of Numerical Analysis

2007 American Control Conference, 2007

Conference on Decision and Control, 2006

This paper presents the design of a vision-based controller for an underactuated, unmanned aerial... more This paper presents the design of a vision-based controller for an underactuated, unmanned aerial vehicle (UAV) equipped with a pan-tilt camera unit (PTCU) to achieve the objective of following a leader vehicle autonomously. The relative position and orientation information is obtained from the monocular camera utilizing homography-based techniques. The proposed controller, built upon Lyapunov design methods, achieves uniform ultimate bounded (UUB) tracking

Visual Communications and Image Processing 2010, 2010

The commercial success and acceptability of 3D technology will critically depend on the overall v... more The commercial success and acceptability of 3D technology will critically depend on the overall visual quality of the rendered images. Therefore Depth Image Based Rendering (DIBR) is a crucial component of the 3D system chain. In this paper, we describe a high quality DIBR system ...

2012 IEEE Consumer Communications and Networking Conference (CCNC), 2012

2010 3DTV-Conference: The True Vision - Capture, Transmission and Display of 3D Video, 2010

The commercial success and acceptability of 3D technol-ogy will critically depend on the overall ... more The commercial success and acceptability of 3D technol-ogy will critically depend on the overall visual quality of the rendered images. Therefore, view synthesis techniques like Depth Image Based Rendering (DIBR) are crucial compo-nents of the 3D system chain. In this paper, ...

2010 IEEE International Conference on Multimedia and Expo, 2010

For future applications like 3DTV and free view point TV, the video data format is likely to comp... more For future applications like 3DTV and free view point TV, the video data format is likely to comprise of multiple views plus pre-estimated geometric information such as a depth-map corresponding to each view. It has been well studied that the depth-maps are useful both from ...

International Conference on Acoustics, Speech, and Signal Processing, 2011

Abstract We present a robust method to map detected facial Action Units (AUs) to six basic emotio... more Abstract We present a robust method to map detected facial Action Units (AUs) to six basic emotions. Automatic AU recognition is prone to errors due to illumination, tracking failures and occlusions. Hence, traditional rule based methods to map AUs to emotions are very ...

... Kiran Nanjunda Iyer, Madhusoodhana Chari, Hariprasad Kannan Samsung India Software Operations... more ... Kiran Nanjunda Iyer, Madhusoodhana Chari, Hariprasad Kannan Samsung India Software Operations Pvt Ltd Bagmane Tech Park, CV Raman Nagar, Bangalore - 560 093 {kiran.iyer, madhu.chari, h.kannan}@samsung.com Abstract ...

Solving optimization tasks based on functions and losses with a topological flavor is a very acti... more Solving optimization tasks based on functions and losses with a topological flavor is a very active and growing field of research in Topological Data Analysis, with plenty of applications in non-convex optimization, statistics and machine learning. All of these methods rely on the fact that most of the topological constructions are actually stratifiable and differentiable almost everywhere. However, the corresponding gradient and associated code is always anchored to a specific application and/or topological construction, and do not come with theoretical guarantees. In this article, we study the differentiability of a general functional associated with the most common topological construction, that is, the persistence map, and we prove a convergence result of stochastic subgradient descent for such a functional. This result encompasses all the constructions and applications for topological optimization in the literature, and comes with code that is easy to handle and mix with other ...