Milind Padalkar

Milind Padalkar
Istituto Italiano di Tecnologia | IIT · Department of Pattern Analysis and Computer Vision

Ph.D.

About

25
Publications
4,527
Reads
How we measure 'reads'
A 'read' is counted each time someone views a publication summary (such as the title, abstract, and list of authors), clicks on a figure, or views or downloads the full-text. Learn more
27
Citations
Introduction
I am a postdoc at Pattern Analysis and Computer Vision (PAVIS) Research Line, Istituto Italiano di Tecnologia (IIT), Genova, Italy. Prior to joining PAVIS, I was working as a senior research engineer at Vehant Technologies. My Ph.D. thesis is titled "Novel Techniques for Auto-inpainting in Heritage Reconstruction", for which I worked under the supervision of Prof. M. V. Joshi at Dhirubhai Ambani Institute of Information and Communication Technology (DA-IICT), Gandhinagar, India. At DA-IICT, I have also worked as teaching assistant and as a junior research fellow (JRF) in the Indian Digital Heritage (IDH)-Hampi project. I completed M.Tech. in Computer Engineering under the supervision of Prof. M. A. Zaveri at S. V. National Institute of Technology (SVNIT), Surat, India in 2010 and B.E. in
Additional affiliations
March 2017 - present
Vehant Technologies Pvt. Ltd.
Position
  • Senior Researcher
December 2011 - April 2017
Dhirubhai Ambani Institute of Information and Communication Technology
Position
  • PhD Student
August 2011 - March 2016
Dhirubhai Ambani Institute of Information and Communication Technology
Position
  • JRF in Indian Digital Hampi Project
Education
July 2004 - June 2008
Finolex Academy of Management and Technology (University of Mumbai)
Field of study
  • Information Technology

Publications

Publications (25)
Preprint
Full-text available
This paper presents a novel setup for automatic visual inspection of cracks in ceramic tile as well as studies the effect of various classifiers and height-varying illumination conditions for this task. The intuition behind this setup is that cracks can be better visualized under specific lighting conditions than others. Our setup, which is designe...
Chapter
Historical monuments are considered as one of the key aspects of our cultural heritage. Unfortunately, due to a variety of factors, the monuments get damaged. The need for preservation of cultural heritage has desiderated research on digitally repairing the photographs of damaged monuments. One may think of digitally undoing the damage to the monum...
Chapter
In Chapters 4-6 we discussed inpainting techniques wherein the regions to be inpainted are manually provided by the users. When looking at heritage monuments, especially statues, there is a consensus about the desire to view these without any damage to the dominant facial regions. This encourages the exploration for a technique that automatically d...
Chapter
Digitized 3D models of heritage scenes/objects can be generated with the help of data captured using laser scanners or a large number of photographs. Due to self-occlusion or difficulty in capturing the scene/object from a particular viewpoint, some parts of the scene/object may not be captured at all. It may also happen that the monument is ruined...
Chapter
Enhancement of spatial resolution is a preliminary step for digital reconstruction of heritage sites as this enables the viewers to perceive finer details of the skillful historic work. This chapter presents a super-resolution approach for resolution enhancement that exploits the recurrence of the visual content in natural images, which is an impor...
Chapter
In Chapters 2 and 3 we discussed techniques for super-resolution, while in Chapter 4, we pointed out the limitations of the existing inpainting approaches and discussed an autoregressive modelbased technique to overcome the limitations. Nevertheless, the technique proposed in Chapter 4 did not deal with the problem of finding good exemplar patches...
Chapter
In Chapter 7 we discussed a technique for automatic detection and inpainting of the damaged eye, nose and lip regions in facial images of statues. In this chapter, we consider the damage in non-facial regions and describe techniques to automatically detect these and inpaint them. In particular, we consider the non-facial region that can have damage...
Chapter
Self-learning, introduced in the previous chapter, uses only one image to achieve super-resolution with high magnification factors. But, as the image resolution increases, the number of patches in the dictionary also increases dramatically, and makes the SR reconstruction computationally prohibitive. It employs l1-minimization to exploit the sparsi...
Chapter
In Chapter 4 we discussed an image inpainting technique that fills the missing pixels by using the pixel-neighborhood relationship in exemplars, while in Chapter 5 we discussed various attempts to improve inpainting. Based on our observations from Chapter 5, in this chapter we discuss a method that not only inpaints the given missing region but als...
Chapter
In this chapter, we discuss the challenges and future directions involved in using super-resolution and inpainting for digital heritage reconstruction. The chapter provides brief insights of these challenges, and is meant to serve as fodder that will inspire the readers to pursue them in the future.
Book
Heritage sites across the world have witnessed a number of natural calamities, sabotage and damage from visitors, resulting in their present ruined condition. Many sites are now restricted to reduce the risk of further damage. Yet these masterpieces are significant cultural icons and critical markers of past civilizations that future generations ne...
Conference Paper
Full-text available
Past two decades have seen significant advancement in the techniques for scene completion and image super-resolution. Although many of the approaches solve these two problems by searching and processing of similar patches for estimating the unknown pixel values, the two problems have been addressed independently. In applications like creating immer...
Article
Full-text available
The need for preservation of cultural heritage has necessitated the research on digitally repairing the photographs of damaged monuments. In this paper, we first propose a technique for automatically detecting the cracked regions in photographs of monuments. Unlike the usual practice of manually selecting the mask for inpainting, the detected regio...
Conference Paper
Full-text available
Historical monuments are considered as one of the key aspects for modern communities. Unfortunately, due to a variety of factors the monuments get damaged. One may think of digitally undoing the damage to the monuments by inpainting, a process to fill-in missing regions in an image. A majority of inpainting techniques reported in the literature requ...
Conference Paper
Full-text available
We are often required to retouch images in order to improve their visual appearance, by removing the visual discontinuities like breaks and damaged regions. Such retouching of images may be achieved by inpainting. Current techniques for image inpainting require the user to manually select the target regions to be inpainted. Very few techniques for...
Conference Paper
Full-text available
In this paper, we propose an iterative exemplar based inpainting technique. A set of exemplars is automatically searched considering a window around every pixel in the region to be inpainted. The novelty of our technique lies in the use of parameters of an autoregressive (AR) model that are estimated using the non-negatively constrained least squar...
Thesis
Full-text available
Videos have become a popular means of entertainment over the years. With the increase in the amount of user generated videos, a large collection is readily available on popular video sharing websites. Searching for videos with desired content from such a large collection is becoming a tedious task. The viewers require better control over the video...
Conference Paper
Full-text available
Today, with the availability of large amount of video files, searching for videos with desired content is becoming a tedious task. The viewers require better control over the video data and for this reason the video browsing and indexing applications are being developed. These applications are required to segment the video into shots in the initial...

Network

Cited By

Projects

Project (1)