Sports-10K and TV Series-1M Video Datasets

[Project page]


videostrfig


 

About

We introduce two large video datasets namely Sports-10K and TV series-1M to demonstrate scene text retrieval in the context of video sequences. The first one is from sports video clips, containing many advertisement signboards, and the second is from four popular TV series: Friends, Buffy, Mr. Bean, and Open All Hours. The TV series-1M contains more than 1 million frames. Words such as central, perk, pickles, news, SLW27R (a car number) frequently appear in the TV series-1M dataset. All the image frames extracted from this dataset are manually annotated with the query text they may contain. Annotations are done by a team of three people for about 150 man-hours. We use 10 and 20 query words to demonstrate the retrieval performance on the Sports-10K and the TV series-1M datasets respectively.


Downloads

Please mail us at This email address is being protected from spambots. You need JavaScript enabled to view it. for a copy of these two datasets (To be used for research purpose only).


Publications

Anand Mishra, Karteek Alahari and C. V. Jawahar.
Image Retrieval using Textual Cues
ICCV 2013 [PDF]


Bibtex

If you use this dataset, please cite:

@InProceedings{MishraICCV13,
  author    = "Mishra, A. and Alahari, K. and Jawahar, C.~V.",
  title     = "Image Retrieval using Textual Cues",
  booktitle = "ICCV",
  year      = "2013",
}

Related datasets


Contact

For any queries about the dataset feel free to contact Anand Mishra. Email:This email address is being protected from spambots. You need JavaScript enabled to view it.

 

The IIIT Scene Text Retrieval (STR) Dataset

[Project Page]

strfig


About

The IIIT STR dataset is harvested from Google and Flickr image search. Query words like coffee shop, motel, post office, high school, department were used to collect the images. Additionally, query words like sky, building were used in Flickr to collect some random distractors (images not containg text). The dataset contains 10,000 images in all. The images are manually annotated to say whether they contain a query word or not. Annotation for all the 50 query words used in our paper is available. Each query word appears 10-50 times in the dataset.


Downloads

IIIT STR (758 MB)
README


Publications

Anand Mishra, Karteek Alahari and C. V. Jawahar.
Image Retrieval using Textual Cues
ICCV 2013 [PDF]


Bibtex

If you use this dataset, please cite:

@InProceedings{MishraICCV13,
  author    = "Mishra, A. and Alahari, K. and Jawahar, C.~V.",
  title     = "Image Retrieval using Textual Cues",
  booktitle = "ICCV",
  year      = "2013",
}

Related datasets


Contact

For any queries about the dataset feel free to contact Anand Mishra. Email:This email address is being protected from spambots. You need JavaScript enabled to view it.

 

IIIT-Seed dataset


IIIT-Seed (Semantic Edges) dataset contains 500 RGBD images of varying complexity. This dataset consists of objects such as tables, chairs, cupboard shelves, boxes and household objects in addition to walls and floors. This dataset is reported in the paper cited below :

Nishit Soni, Anoop M. Namboodiri, CV Jawahar and Srikumar Ramalingam. Semantic Classification of Boundaries of an RGBD Image. In Proceedings of the British Machine Vision Conference (BMVC 2015), pages 114.1-114.12. BMVA Press, September 2015. [paper] [abstract] [poster] [code] [bibtex]

  • Download the dataset : here. Cite our paper if you use this dataset. [bibtex].
  • Test and train files : test split and train split.
  • Download 100 images of NYU dataset used to test our approach on: here.
  • Pb edge links for our dataset as well as the NYU dataset : here.
  • Groundtruth for both the datasets : here.
  • readme.txt

Image Retrieval using Textual Cues

Anand Mishra, Karteek Alahari and C V Jawahar

ICCV 2013


Abstract

We present an approach for the text-to-image retrieval problem based on textual content present in images. Given the recent developments in understanding text in images, an appealing approach to address this problem is to localize and recognize the text, and then query the database, as in a text retrieval problem. We show that such an approach, despite being based on state-of-the-art methods, is insufficient, and propose a method, where we do not rely on an exact localization and recognition pipeline. We take a query-driven search approach, where we find approximate locations of characters in the text query, and then impose spatial constraints to generate a ranked list of images in the database. The retrieval performance is evaluated on public scene text datasets as well as three large datasets, namely IIIT scene text retrieval, Sports-10K and TV series-1M, we introduce.


Contributions

  • Query driven approach for scene text based image retrieval
  • Scene text indexing without explicit text localization and recognition
  • Category as well as instance retrieval
  • The largest datasets for the problem

Paper

ICCV 2013 Paper / Poster


Datasets

IIIT Scene Text Retrieval (IIIT STR)

Video Scene Text Retrieval Datasets (TV series-1M and Sports-10K)


Extended results

TBA 


BibTeX

@InProceedings{Mishra13,
  author    = "Mishra, A. and Alahari, K. and Jawahar, C.~V.",
  title     = "Image Retrieval using Textual Cues",
  booktitle = "Proceedings of IEEE International Conference on Computer Vision",
  year      = "2013"
}

Acknowledgements

Anand Mishra is partly supported by MSR India PhD Fellowship 2012.


Copyright Notice

The documents contained in these directories are included by the contributing authors as a means to ensure timely dissemination of scholarly and technical work on a non-commercial basis. Copyright and all rights therein are maintained by the authors or by other copyright holders, notwithstanding that they have offered their works here electronically. It is understood that all persons copying this information will adhere to the terms and constraints invoked by each author's copyright.

Culling an Object Hierarchy to a Frustum Hierarchy


Introduction

Visibility culling of a scene is central to any interactive graphics application. The idea is to limit the geometry sent down the rendering pipeline to only the geometry with a fair chance of finally becoming visible. It is important for the culling stage to be fast for it to be effective; otherwise the performance gain achieved will be overshadowed. Hierarchical scene structures are commonly used to speed up the process. Hierarchical culling of bounding boxes to a view frustum is fast and sufficient in most applications. However, when there are multiple view frustums (as in a tiled display wall), visibility culling time becomes substantial and cannot be hidden by pipelining it with other stages of rendering.

Here, we address the problem of culling an object hierarchy to a hierarchically organized set of frustums, such as those found in tiled displays and shadow volume computation. We present an adaptive algorithm to unfold the Object Hierarchy (OH) or Frustum hierarchy (FH) at every stage in the culling procedure. Our algorithm computes from-point visibility and is conservative. The precomputation required is minimal, allowing our approach to be applied for dynamic scenes as well.


Design & Features

Features of the Adaptive OH and FH culling Algo ::fh2

  • Faster running time than existing solutions based etirely on either OH or FH
  • Use of Heirarchies potentially eliminates half objects and frustums in each step
  • Sublinear running time, adapts according to viewpoint and "goodness" of the scene.
  • Simple PCA based preprocessing required to compute Oriented Bounding boxes
  • Can work with OBB's, AABB's or Bounding Spheres
  • Scales sublineraly to to both object and frustum count, suitable for very large number of objects and large tiled displays
  • Dynamic objects can be used as only cheap OBB/AABB computation is needed every frame
  • Easily extendable to a generalized arrangement of view frustums

We first find all the objects in the Primary view frustum and then call the Adaptive algorithm for every object present inside the Primary view frustum. The Adaptive algoritm is recursive in nature. The basic steps involved are::

Algorithm Adaptive Culling. Inputs: Object Node and Frustum Nodealgo

  • If the Frustum node is a leaf node, mark the Object node visible to this frustum and return.
  • [L,C,R] = Classify the children of the Object Node according to their orientation with respect to the Frusum node's bisection plane.
  • Send All objects in set C to both sub-frustums of the Frustum Node using this algorithm again
  • Send All objects in set L to the left sub-frustum using this algorithm again
  • Send All objects in set R to the right sub-frusrum using this algorithm again

The time complexity of the algorithm is roughly of O (min (N*logM, M*logN)) where N = Number of objects and M = Number of view frustums.


Related Publication

  • Nirnimesh, Pawan Harish and P. J. Narayanan - Garuda: A Scalable, Tiled Display Wall Using Commodity PCs Proc. of IEEE Transactions on Visualization and Computer Graphics(TVCG), Vol.13, no.~5, pp.864-877, 2007. [PDF]

  • Nirnimesh, Pawan Harish and P. J. Narayanan - Culling an Object Hierarchy to a Frustum Hierarchy, 5th Indian Conference on Computer Vision, Graphics and Image Processing, Madurai, India, LNCL 4338 pp.252-263,2006. [PDF]


Associated People