OBJECTS

Table of Contents

OBJECTS

Artificial objects

Datasets that contain images of objects and words and etc. that do not exist in the real world.

Artificial objects in different views:

 

Description: This database has 2 object sets, the “amoeboid” set and the “regular set“. One set contains 62 objects, which the creators call “amoeboid“, each is shown from 5 different angels with 20° difference. 31 of these “amoeboids” have part which you can’t see from all the angels but you can see every part of the other 31 from all angels. The “regular set“ contains 80 objects, 40 of them have parts that can’t be seen from all the angels but the other 40 have parts that can be seen from all 5 positions.

License: Not known.

Link: http://ww2.psy.cuhk.edu.hk/~mael/Stimuli.html

Reference:

Wong, A.C.-N., & Hayward, W.G. (2005). Constraints on view combination: Effects of self-occlusion and difference between familiar views. Journal of Experimental Psychology: Human Perception & Performance, 31, 110-121.

 

Novel Objects and Unusual Name Database (NOUN)

 

Description: This set contains 64 made up objects which do not consist in reality but look like real objects at the same time. This set also contains pseudo words, which are words that sound like real English words but don‘t exist, and novel nouns.

License: Creative Commons Attribution NonCommercial

Link: http://michaelhout.com/?page_id=759

Reference:

Horst, J. S., & Hout, M. C. (in press). The Novel Object and Unusual Name (NOUN) Database: A collection of novel images for use in experimental research. Behavior Research Methods.

Fékk leyfir frá jessica(Replace this parenthesis with the @ sign)sussex.ac.uk

Ziggerins:

Description: this database contains pictures of isolated unreal objects called “ziggerins” with white background. There are about 6 categories of “ziggerins” each has 12 “ziggerins” which have “similar basic structure but differ in style or the detailed manifestations of the parts.  “The 12 styles are formed by variations of part along three dimensions: 1) cross section shape, 2) cross-section size change and 3) aspect ratio of the cross section.”

License: Not known.

Link: http://ww2.psy.cuhk.edu.hk/~mael/Stimuli.html

References:.

Wong, A. C.-N., Palmeri, T.J., & Gauthier, I. (2009). Conditions for face-like expertise with objects: Becoming a Ziggerin expert – but which type? Psychological Science, 20(9), 1108-1117.

 

Real Objects

Images of diverse real objects.

Bank of Standardized Stimuli (BOSS):

Description: The BOSS set contains 480 pictures of various real life objects. Each object is shown isolated and not part of any scene but with a white background. The objects are both man-made and natural like vegetables.

License: https://creativecommons.org/licenses/by-sa/3.0/

Link: https://sites.google.com/site/mathieubrodeur/Home/boss

Reference: Brodeur, M. B., Dionne-Dostie, E., Montreuil, T., & Lepage, M. (2010). The bank of standardized stimuli (BOSS), a new set of 480 normative photos of objects to be used as visual stimuli in cognitive research. PloS ONE, 5(5), e10773 https://docs.google.com/viewer?a=v&pid=sites&srcid=ZGVmYXVsdGRvbWFpbnxtYXRoaWV1YnJvZGV1cnxneDo1MGE3Yzg0MGEyODhkY2Mz

Big and small objects:

Set 1

Description: 400 pictures of big and small real objects. The objects are not part of a scene but shown isolated with white background.  These objects can be found in nature, as in plants and fruits, but also man made but what they seem to have in common is that most of us have seen these objects before in our daily lives. The objects are categorized in two groups, small objects like glasses and big objects like cars.

License: Not known

Link: http://konklab.fas.harvard.edu/#

Reference:

Konkle, T., Olivia, A. (2012). A Real_World Size Organization of Object Responses in Occipitotemporal Cortex. Neuron, colume 74( 6), 1114-1124. http://www.sciencedirect.com/science/article/pii/S0896627312004412

 

Set 2

Description: 120 pictures of big and small real objects. The objects are not part of a scene but shown isolated with white background. These objects can be found in nature, as in plants and fruits, but also man made but what they seem to have in common is that most of us have seen these objects before in our daily lives. The objects are categorized in to two groups, small objects like glasses and big objects like cars.

License: Not known.

Link: http://konklab.fas.harvard.edu/#

Reference:  Caramazza, A., Konkle, T.(2013). Tripartite Organization of the Ventral Stream by Animacy and Object Size. The Journal of Neuroscience, 33(25), 10235-10242.

https://dash.harvard.edu/bitstream/handle/1/12362624/1781465.pdf?sequence=1

 

Black and white images of objects

Description: This set contains 299 black and white images of real life stuff from a wallet to an aerial. The images are not part of a scene but black drawings on a white background. They are ordered alphabetically both in English and in French.

License: Not known.

Link: http://leadserv.u-bourgogne.fr/bases/pictures/

References: Bonin, P., Peereman, R., Malardier, N., Méot, A., & Chalard, M. (2003). A new set of 299 pictures for psycholinguistic studies: French norms for name agreement, image agreement, conceptual familiarity, visual complexity, image variability, age of acquisition, and naming latencies. Behavior Research Methods, Instruments, & Computers, 35(1), 158-167.

 

Caltech 101:

Description: This database contains images of objects from 101 different classes with 40-800 pictures in each class. Most of these classes contain around 50 pictures. Some of these pictures show objects which are a part of a scene but others show them isolated with white or black background.

License: Not known.

Link: http://www.vision.caltech.edu/Image_Datasets/Caltech101/Caltech101.html

References: If you only use their images then cite:

L.Fei-Fei, R. Fergur adn P. Perona. Learning generative visual models from few treining examples: an incremental Beyesian approuch tested on 101 object categories. IEEE. CVPR 2004, Workshop on Generative-Model Based Vision. 2004.

If you are using their images and annotations:

L.Fei-Fei, R. Fergus and P. Perona. One-shot learning of object catagories. IEEE Trans. Pattern Recognition and Machine Intelligense. In press.

 

 

Caltech 256 database:

Description: This dataset contains 30607 pictures of objects from 256 different classes both photos of real life objects and drawings of them. Some of the pictures show isolated objects which are not a part of any scenes but other images show objects as a part of a scene. There are at least 80 pictures or more in each category according to their online slide show.

License: Not known.

Link: http://www.vision.caltech.edu/Image_Datasets/Caltech256/

References: Griffin, G. Holub, AD. Perona, P. The Caltech 256. Caltech Technical Report.

 

3D Objects on Turntable:

Description: This large set has photos, of various object on an automated turntable with a white background, taken from 144 different angels with 5° difference under 3 different lighting conditions.  The objects are photographed by 2 cameras.

License: Not know.

Link: http://www.vision.caltech.edu/pmoreels/Datasets/TurntableObjects/index.html

References:

  1. Moreels, P., & Perona, P. (2005, October). Evaluation of features detectors and descriptors based on 3D objects. In Tenth IEEE International Conference on Computer Vision, 1, 800-807.
  2. Moreels, P., & Perona, P. (2007). Evaluation of features detectors and descriptors based on 3d objects. International Journal of Computer Vision, 73(3), 263-284.

 

 

 

100 Exemplar Pairs:

Description: 100 pictures. Each showing 2 object which would be categorized as the same but have a different look. The objects are not shown as a part of a scene but with a white background.

License: Not known.

Link: http://bradylab.ucsd.edu/stimuli.html

Reference:

Brady, T. F., Konkle, T., Alvarez, G. A. and Oliva, A. (2008). Visual long-term memory has a massive storage capacity for object details. Proceedings of the National Academy of Sciences, USA, 105 (38), 14325-14329. http://www.pnas.org/content/105/38/14325.full

 

 

Giuseppe Toys Dataset:

Description: This database contains photos categorized in 4 subsets. The first set, called “Training images“, has 1 to 4 picture, 61 pictures in all, of each object (mostly stuffed animals and toy vehicles). The second subset, called “Test-scenes“, has 52 picture of various scenes with many toys from the training set. The third subset, “Test-single toy“, has 62 photos of which show only one toy from the training set. The last subset, “Test-not toy“, has 26 pictures that do not show any toys from the training set.

License: Not known.

Link: http://www.vision.caltech.edu/pmoreels/Datasets/Giuseppe_Toys_03/#Description

Reference: Not known.

 

Home Objects:

Description: This database contains 224 photos of various objects from the someone‘s home. These objects are used in the kitchen, bathroom or living room. They are not shown isolated and are a part of a scene and some pictures contain more than one object.

License: Not known.

Link: http://www.vision.caltech.edu/pmoreels/Datasets/Home_Objects_06/

Reference: Not known.

 

 

“Massive Memory” Unique Object Images:

Description: contains 2400 pictures of various objects. It also contains 200 pictures which show 2 different objects from the same category and 200 images of the same object twice but in different pose.

License: not known.

Link: http://konklab.fas.harvard.edu/#

References: Brady, T. F., Konkle, T., Alvarez, G. A., & Oliva, A. (2008). Visual long-term memory has a massive storage capacity for object details. Proceedings of the National Academy of Sciences, 105(38), 14325-14329.

 

“Massive Memory” Object Categories:

Description: this database has three subsets. The first includes 200 different classes with 17 pictures of different subtypes in each. The second contains pictures of 240 classes from one to sixteen images of different objects which look the same from the same class. This database also provides information about the different perceptual knowledge and conceptual knowledge each object has.

License: not known

Link: http://konklab.fas.harvard.edu/#  also on this website http://bradylab.ucsd.edu/stimuli.html

References: Brady, T. F., Konkle, T., Alvarez, G. A., & Oliva, A. (2010). Conceptual Distincitveness Supports Detailed Visual Long-Term Memory for Real-World Objects. Journal of Experimental Psychology: General 2010, 139(3), 558-578.

 

540 Objects designed to be color-rotated, plus code to rotate them:

Description: “this set contains 540 objects with a code to change their colour and turn them so you can see them from various angles.”

License: Not known

Link: http://bradylab.ucsd.edu/stimuli.html

Reference:

Brady, T. F., Konkle, T.F., Gill, J., Oliva, A. and Alvarez, G.A. (2013). Visual long-term memory has the same limit on fidelity as visual working memory. Psychological Science, 24(6), 981-990.

 

 

 

Object quartets: State x Exemplar and State x Color:

Description: This set shows 100 pictures which show two objects which have different colour or type and each shown in a different condition. For an example you see two pens who either look alike except one is red but the other is blue or they have the same colour but are different type of pens, and you see them with their lid on and off.

License: not known

Link: http://konklab.fas.harvard.edu/#

Reference: Brady, T. F., Konkle, T., Gill, J., Oliva, A., & Alvarez, G. A. (2012). Long-term memory has the same limit on fidelity as working memory. Manuscript submitted for publication.

 

 

Objects from distinct categories:

Description: This set has 2400 images of various real objects from toys, birds to airplanes. The objects are isolated in each picture and not a part of a scene but with a white background. There are 160 kinds of objects and around 15 pictures of each object. The pictures are organized alphabetically in the set.

License: Not known.

Link: http://konklab.fas.harvard.edu/#

Reference:

Brady, T. F., Konkle, T., Alvarez, G. A., Oliva, A. (2008). Visual long-term memory has a massive storage capacity for object details. PNAS, 105(38), 14325-14329.

 

Object Size Range:

Description: 100 images showing real life objects from buildings, tools to toys with white, plain, background.  Along with the images there are information about the objects’ sizes in real life considering the other objects. The objects where ordered from the smallest to largest and including the information about each object there are information about their rank.

License: Not known.

Link: http://konklab.fas.harvard.edu/#

Reference: Konkle, T., & Oliva, A. (2011). Canonical visual size for real-world objects. Journal of experimental psychology: human perception and performance, 37(1), 23 -37.

 

Object Size Stroop:

Description: this database contains 400 real world objects, 200 of them are large in real life but the other 200 are small. They are all in front of a white background. The pictures are all in the same sizes but the objects fill into the image at a different degree. According to Konkle, Olivia (2012) the images each image of a small and a large object where shown together so that the small object looked bigger than the large ones or the other way around.

License: Not known.

Link: http://konklab.fas.harvard.edu/#

Reference:

  1. Konkle, T., & Oliva, A. (2012). A familiar-size Stroop effect: real-world size is an automatic property of object representation. Journal of Experimental Psychology: Human Perception and Performance, 38(3), 561.
  2. Konkle, T., & Oliva, A. (2012). A real-world size organization of object responses in occipitotemporal cortex. Neuron, 74(6), 1114-1124.

 

100 Objects in 2 states x 2 exemplars:

Description: This set contains 400 pictures in all of 100 different objects. In each picture one object is shown isolated, that is it is not part of any scene, with white background. There are 2 different examples of each object and 2 photos of the same object but in different state or position. For example one category is of books with 4 different pictures. 2 of those pictures are of the same book but in one picture the book is open in another image the book is closed. The other two pictures in the same category are of another book, one picture of the book open the other closed and so forth.

License: Not known.

Link: http://bradylab.ucsd.edu/stimuli.html

Reference:

Brady, T. F., Konkle, T., Alvarez, G.A., and Oliva, A. (2013). Real-world objects are not represented as bound units: Independent forgetting of different object details from visual memory. Journal of Experimental Psychology: General, 142(3), 791-808.

 

 

100 Objects in 2 states x 2 colors:

Description: This set contains 400 pictures in all of 100 different objects. The objects are shown isolated with a white background. There are 4 picture of each object, in 2 of the pictures the object is in the same colour but in different state and the other 2 are of the same object in another colour in 2 different states. For example there are 2 pictures of the same black and blue bag. In one photo the bag is closed but in the other one the bag is open. Then there is another image of a bag which looks just the same but is yellow and black and in one picture that same bag is open but in another on the bag is closed.

License: Not known.

Link: http://bradylab.ucsd.edu/stimuli.html

Reference: Brady, T. F., Konkle, T., Alvarez, G.A., and Oliva, A. (2013). Real-world objects are not represented as bound units: Independent forgetting of different object details from visual memory. Journal of Experimental Psychology: General, 142(3), 791-808.

 

 

Pairs of Objects differing at the exemplar level

Description: 200 pictures each of 2 isolated objects, with a white background, which are similar and from same basic category but obviously different at the same time. For an example one picture might contain 2 mirrors which have different frames or shape.

License: Not known.

Link: http://konklab.fas.harvard.edu/#

Reference:

Brady, T. F., Konkle, T., Alvarez, G. A., Oliva, A. (2008). Visual long-term memory has a massive storage capacity for object details. PNAS,  105(38), 14325-14329.

 

Pairs of object, differing in state or pose:

Description: 200 pictures. Each picture shows the same isolated object twice but from different angels. For an example one picture might show you the same chair from two different angels.

License: Not known.

Link: http://konklab.fas.harvard.edu/#

Reference: Brady, T. F., Konkle, T., Alvarez, G. A., Oliva, A. (2008). Visual long-term memory has a massive storage capacity for object details. PNAS,  105(38), 14325-14329.

 

Poporo image set:

Description: This database has 800 pictures of pair of objects. The objects are not shown as a part of any scene but isolated with a white background.  These are various real objects from animals, plants to vehicles. This set does not include human faces, letters or other symbols. These object pairs are either rated as “semantically related” or unrelated.

License: Not known.

Link: http://www.oszillab.net/downloads.php

Reference: Kovalenko, L.Y., Chaumon, M., & Busch, N.A. (2012). A pool of pairs of related objects (POPORO) for investigating visual semantic integration: Behavioral and electrophysiological validation. Brain Topography, 25(3), 272-284. http://link.springer.com/article/10.1007%2Fs10548-011-0216-8

 

Quality Colour Images with Norms from Seven Psycholinguistic Variables:

Description: This set has 360 pictures in all with 120 coloured pictures in three categories, “Living things“, “Non-living things“, “Nature“. The objects are shown isolated and not a part of any scenery with a white background. Each category has couple of subcategories with pictures in each subcategory. The “Living things” category contains: animals, insects, vegetables, nuts, trees, marine creatures, fruits, body parts, flowers and bird. The “Non-living things” contains musical instruments, buildings, jewellery, food, tools, vehicles, weapons, desk mat, sport games, kitchen tool, clothing and furniture. The “Nature” category contains: mountain, volcano, wave, cliff, waterfall, ice, iceberg, island, stone, gold, coal, puddle, moon, sun, sea and a cloud.

License: Creative Commons Attribution

Link: http://dx.plos.org/10.1371/journal.pone.0037527 (see Supporting information for download links)

Reference:

Moreno-Martínez, F. J., & Montoro, P. R. (2012). An ecological alternative to Snodgrass & Vanderwart: 360 high quality colour images with norms for seven psycholinguistic variables. PLoS One, 7(5), e37528

Segmentation evaluation database:

Description: This database contains 2 different set which contain around 200 black and white images of various objects. One dataset contains images which show one object but the other shows two objects. The objects are all a part of a scene. The creators of this database provided a code for the evaluation of the given segmentation algorithm.

License: Depends on the image see website.

Link: http://www.wisdom.weizmann.ac.il/~vision/Seg_Evaluation_DB

Reference: Not known

Similar Object and Lure Image Database (SOLID):

Description:  201 categories of grayscale objects, with approximately 17 exemplars per set. SOLID offers both a large number of stimuli and a considerable range of similarity levels.

License: See link.

Link: https://www.click2go.umip.com/i/copyright/solid.html

Reference: Frank, D., Gray, O., & Montaldi, D. (2019). SOLID-Similar object and lure image database. Behavior Research Methods, 1-11. https://link.springer.com/article/10.3758/s13428-019-01211-7

 

100 State Pairs:

Description: 100 pictures of pictures each picture shows the same objects twice in different conditions for an example a picture of a CD player opened and closed. The objects are shown isolated with white background.

License: Not known.

Link: http://bradylab.ucsd.edu/stimuli.html

Reference:

Brady, T. F., Konkle, T., Alvarez, G. A. and Oliva, A. (2008). Visual long-term memory has a massive storage capacity for object details. Proceedings of the National Academy of Sciences, USA, 105 (38), 14325-14329. http://www.pnas.org/content/105/38/14325.full

 

2400 Unique objects:

Description: 2400 images of isolated various objects with a white background.

License: Not known.

Link: http://bradylab.ucsd.edu/stimuli.html

Reference:

Brady, T. F., Konkle, T., Alvarez, G. A. and Oliva, A. (2008). Visual long-term memory has a massive storage capacity for object details. Proceedings of the National Academy of Sciences, USA, 105 (38), 14325-14329. http://www.pnas.org/content/105/38/14325.full

 

Objects that move

STL-10 dataset:

 

Description: this database contains 10 categories of animals and moving vehicles. The categories are airplane, truck, car, ship and monkey, horse, dog, deer, cat and bird 500 images for training and 800 test pictures in each category. The images are 96×96 pixels and in colour. This base is used to make a “feature-”, “deep-”, “self-taught” – learning algorithms.

License: Not known.

Link: https://cs.stanford.edu/~acoates/stl10/

Reference:

Adam Coates, Honglak Lee, Andrew Y. Ng An Analysis of Single Layer Networks in Unsupervised Feature Learning AISTATS, 2011.