Observe the picture above. Unless you are already familiar with this picture, in all likelihood it will not make much sense. There are just black patches on a white background.
So the name of this is “Dalmatian dog”, does this help? If not, use the help check boxes, starting with “Help 1”.
Once the dog has been seen, it stays obvious – one-shot learning has taken place.
This example demonstrates the amazing capability of our visual system to organise noisy and ill-defined visual input into interpretable shapes (with sometimes exaggerated results as shown, e.e., by the “face on Mars”.
The movie in “Help 2” further demonstrates how strongly texture segregation (separating the figure from ground) is affected by motion gradients.
[On some browsers the movie does not play, have not yet solved this issue.]
Gregory R (1970) “The intelligent eye” McGraw-Hill, New York (Photographer: Ronald C James)
Gregory R (2001) The Medawar Lecture 2001 Knowledge for vision: vision for knowledge. Phil Trans Biol Sci (image a little smaller there)
First publication of the picture probably in Life Magazine:58;7 1965-02-19, p 120.
For the movie I gratefully thank Wim van de Grind