OTCBVS Benchmark Dataset Collection

OTCBVS

The web address of OTCBVS Benchmark has changed and please update your bookmarks.

Introduction

This is a publicly available benchmark dataset for testing and evaluating novel and state-of-the-art computer vision algorithms. Several researchers and students have requested a benchmark of non-visible (e.g., infrared) images and videos. The benchmark contains videos and images recorded in and beyond the visible spectrum and is available for free to all researchers in the international computer vision communities. Also it will allow a large spectrum of IEEE and SPIE vision conference and workshop participants to explore the benefits of the non-visible spectrum in real-world applications, contribute to the OTCBVS workshop series, and boost this research field significantly.

This benchmark is to be used for educational and research purposes only, and this benchmark must be acknowledged by the users.

Open Call for OTCBVS Benchmark dataset [TXT]

Questions or comments regarding this benchmark should be sent to the benchmark steering committee at guoliang.fan@okstate.edu.


Dataset 01: OSU Thermal Pedestrian Database


Topic of Interest:
Person detection in thermal imagery.

Sensor Details:
Raytheon 300D thermal sensor core
75 mm lens
Camera mounted on rooftop of 8-story building
Gain/focus on manual control

Data Details:
Pedestrian intersection on the Ohio State University campus
Number of sequences = 10
Total number of images = 284
Format of images = 8-bit grayscale bitmap
Image size = 360 x 240 pixels
Sampling rate = non-uniform, less than 30Hz
Environmental information for each sequence provided in subdirectories
Ground truth provided in subdirectories as list of bounding boxes (with approximately same aspect ratio) around people.
For the ground truth data, we selected only those people that were at least 50% visible in the image (i.e., highly occluded people were not selected).

Requested Citation Acknowledgment:
IEEE OTCBVS WS Series Bench; J. Davis and M. Keck, "A two-stage approach to person detection in thermal imagery," In Proc. Workshop on Applications of Computer Vision, January 2005 [pdf]

Point-of-contact:
James W. Davis, jwdavis[at]cse.ohio-state.edu

Download:
Click here to download this dataset.


Dataset 02: IRIS Thermal/Visible Face Database


Topic of Interest:
Simultaneously acquired unregistered thermal and visible face images under variable illuminations, expressions, and poses.

Sensor Details:
Thermal - Raytheon Palm-IR-Pro
Visible - Panasonic WV-CP234
Setup:

Data Details:
Total size of 1.83 GB
Image size: 320 x 240 pixels (visible and thermal)
4228 pairs of thermal and visible images
176-250 images/person, 11 images per rotation (poses for each expression and each illumination)
30 individuals - Expression, pose, and illumination
Expression: ex1, ex2, ex3 - surprised, laughing, angry (varying poses)
Illumination: Lon (left light on), Ron (right light on), 2on (both lights on), dark (dark room), off (left and right lights off), varying poses

Requested Citation Acknowledgment:
IEEE OTCBVS WS Series Bench; DOE University Research Program in Robotics under grant DOE-DE-FG02-86NE37968; DOD/TACOM/NAC/ARC Program under grant R01-1344-18; FAA/NSSA grant R01-1344-48/49; Office of Naval Research under grant #N000143010022.

Point of Contact:
Besma Abidi, besma[at]utk.edu

Download:
Click here to download this dataset.


Dataset 03: OSU Color-Thermal Database


Topic of Interest:
Fusion of color and thermal imagery,
Fusion-based object detection in color and thermal imagery

Sensor Details:
Thermal Sensor: Raytheon PalmIR 250D, 25 mm lens
Color Sensor: Sony TRV87 Handycam

Cameras mounted adjacent to each other on tripod at two locations approximately 3 stories above ground
Gain/focus on manual control

Data Details:
Busy pathway intersections on the Ohio State University campus
Number of color/thermal sequences = 6 (3 at each location)
Total number of images = 17089
Format of images = Thermal: 8-bit grayscale bitmap, Color: 24-bit color bitmap
Image size = 320 x 240 pixels
Sampling rate = approx. 30Hz
Color/Thermal images registered using homography with manually-selected points
Files containing tracking results on the dataset are provided by Alex Leykin

Requested Citation Acknowledgment:
IEEE OTCBVS WS Series Bench; J. Davis and V. Sharma, "Background-Subtraction using Contour-based Fusion of Thermal and Visible Imagery," Computer Vision and Image Understanding, Vol 106, No. 2-3, 2007, pp. 162-182.

Point-of-contact:
James W. Davis, jwdavis[at]cse.ohio-state.edu

Download:
Click here to download this dataset.


Dataset 04: Terravic Facial IR Database


Topic of Interest:
Facial analysis with thermal imagery

Sensor Details:
Raytheon L-3 Thermal-Eye 2000AS

Data Details:
Number of thermal sequences = 20
Variations = (front,left,right; indoor/outdoor; glasses, hat)
Format of images = 8-bit grayscale JPEG
Image size = 320 x 240 pixels

Requested Citation Acknowledgment:
IEEE OTCBVS WS Series Bench; Roland Miezianko, Terravic Research Infrared Database.

Point-of-contact:
Roland Miezianko, roland[at]terravic.com

Download:
Click here to download this dataset.


Dataset 05: Terravic Motion IR Database


Topic of Interest:
Detection and tracking with thermal imagery

Sensor Details:
Raytheon L-3 Thermal-Eye 2000AS

Data Details:
Number of thermal sequences = 18 (total)

Format of images = 8-bit grayscale JPEG
Image size = 320 x 240 pixels

Requested Citation Acknowledgment:
IEEE OTCBVS WS Series Bench; Roland Miezianko, Terravic Research Infrared Database.

Point-of-contact:
Roland Miezianko, roland[at]terravic.com

Download:
Click here to download this dataset.


Dataset 06: Terravic Weapon IR Database


Topic of Interest:
Weapon detection and weapon discharge detection with thermal imagery

Sensor Details:
Raytheon L-3 Thermal-Eye 2000AS

Data Details:
Number of thermal sequences = 5 (total)

Format of images = 8-bit grayscale JPEG
Image size = 320 x 240 pixels

Requested Citation Acknowledgment:
IEEE OTCBVS WS Series Bench; Roland Miezianko, Terravic Research Infrared Database.

Point-of-contact:
Roland Miezianko, roland[at]terravic.com

Download:
Click here to download this dataset.


Dataset 07: CBSR NIR Face Dataset


Topic of Interest:
NIR face detection, NIR eye detection, NIR face recognition

Sensor Details:
The images were taken by an NIR camera with active NIR lighting. More details are available in reference below.

Data Details:
3,940 NIR face images of 197 people.
The image size is 480 by 640 pixels, 8 bit, without compression.

Images are divided into a gallery set and a probe set. In the gallery set, there are 8 images per person. In the probe set, 12 images per person. The image information is provided, which gives the image number, person number, and eye coordinates.

Requested Citation Acknowledgment:
IEEE OTCBVS WS Series Bench; Center for Biometrics and Security Research (CBSR) www.cbsr.ia.ac.cn; AuthenMetric Co. Ltd (Beijing) www.authenmetric.com
Also see:
Stan Z. Li, RuFeng Chu, ShengCai Liao, Lun Zhang, "Illumination Invariant Face Recognition Using Near-infrared Images," IEEE Transactions on Pattern Analysis and Machine Intelligence (Special issue on Biometrics: Progress and Directions), Vol.29, No.4, April 2007, pp. 627-639. [pdf]

Point-of-contact:
Stan Z. Li, szli[at]cbsr.ia.ac.cn, szli[at]nlpr.ia.ac.cn

Download:
Click here to download this dataset.


Dataset 08: Audio-Visual Vehicle (AVV) Dataset


Topic of Interest:
Ground level moving vehicle detection and classification under various challenging conditions (occlusions, motion blur, various perspective views).

Sensor Details:
Standoff long distance Laser Doppler Vibrometer (acoustic), Polytech LDV OFV505, HeNe laser 632 nm.
two PTZ cameras, Canon VC-C50i.

Data Details:
961 sets of multimodal vehicles samples from both a local road (25 meters) and a highway (55 meters) locations.
Each set of sample has three files: an audio clip (mono 22.5kHz, 16 bit), an original image shot, and a reconstructed visual image.
Several main categories, bikes, buses, motocycles, 2-door sedan, 4-door sedans, pickup trucks, regular trucks, mini-vans, regular vans, and mixtures.

Requested Citation Acknowledgment:
IEEE OTCBVS WS Series Bench;
Wang, T. and Zhu, Z. (2012) "Real time vehicle detection and reconstruction for improving classification," IEEE Computer Society's Workshop on Applications of Computer Vision (WACV), January 9-11, 2012, Colorado.

Point-of-contact:
Tao Wang, tao.wang[at]baesystems.com
Zhigang Zhu, zhu[at]cs.ccny.cuny.edu

Download:
Click here to download this dataset.