Public Datasets

One of the activities of the VENTURI consortium is to acquire high-quality, carefully labelled datasets. The main goal is to provide a quantitative way to evaluate the performances of the various algorithms developed during the project.

Some of these datasets can be considered of general interest, so we decided to share them with the public.

VENTURI Public Datasets
Title Video Audio Sensors GT Device(s) Notes
Mountain Dataset: outdoor video sequences with GT orientation Yes No Yes Yes Sony Ericsson XPERIA Arc S
Building Dataset: collection of pictures with GT pose Yes No No Yes iPhone4

These datasets were created inside the European funded project VENTURI. You are free to use them, but we would appreciate it if you would include a reference to their source in any derived work and publication you write. Thank you.

 

Links to external public datasets used for testing VENTURI algorithms

External Public Datasets
Title Video Audio Sensors GT Notes
KITTI Vision Benchmark Suite Yes No Yes Yes External dataset (link), used in VENTURI for testing optical flow algorithms
Fifteen Scene Categories, Ponce Research Group Yes No No Yes External dataset (link), used in VENTURI for testing scene classification algorithms
ICDAR 2011: Reading Text in Scene Images Yes No No Yes External dataset (link), for testing text detection and recognition algorithms
KAIST Scene Text Database Yes No No Yes External dataset (link), for testing text detection and recognition algorithms
The Street View Text Dataset Yes No No Yes External dataset (link), for testing text detection and recognition algorithms
  1. No comments yet.
  1. No trackbacks yet.
You must be logged in to post a comment.