Rudolf K. Luneburg, author of Mathematical Theory of Optics (1944), also published the book Mathematical analysis of binocular vision (Princeton UP, 1947) in which he argued that the "psychological space of binocular vision" carries a hyperbolic metric. From the Bull. AMS review:

Some of the topics which are treated in considerable detail are the horopter problem (geodesic lines), the alley problem, and rigid transformations of the hyperbolic visual space. (...) Using the hyperbolic metric, the author calculates the shape of distorted rooms which are congruent to rectangular rooms, that is, rooms with distorted walls and windows which appear (under fixed conditions) to be identical to rectangular rooms with rectangular windows.

In 1959, R. Penrose and independently J. Terrell showed that we won't perceive rapidly moving objects as Lorentz contracted but as rotated. Form V. Weisskopf's Physics Today review:

James Terrell (...) does away with an old prejudice held by practically all of us. We all believed that, according to special relativity, an object in motion appears to be contracted in the direction of motion by a factor $[1-(v/c)^2]^{1/2}$. A passenger in a fast space ship, looking out of the window, so it seemed to us, would see spherical objects contracted to ellipsoids. This is definitely not so according to Terrell's considerations, which for the special case of a sphere were also carried out by R. Penrose. The reason is quite simple. When we see or photograph an object, we record light quanta emitted by the object when they arrive simultaneously at the retina or at the photographic film. This implies that these light quanta have not been emitted simultaneously by all points of the object. (...) In special relativity, this distortion has the remarkable effect of canceling the Lorentz contraction so that objects appear undistorted but only rotated. This is exactly true only for objects which subtend a small solid angle.

Rudolf K. Luneburg, author of Mathematical Theory of Optics (1944), also published the book Mathematical analysis of binocular vision (Princeton UP, 1947) in which he argued that the "psychological space of binocular vision" carries a hyperbolic metric. From the Bull. AMS review:

Some of the topics which are treated in considerable detail are the horopter problem (geodesic lines), the alley problem, and rigid transformations of the hyperbolic visual space. (...) Using the hyperbolic metric, the author calculates the shape of distorted rooms which are congruent to rectangular rooms, that is, rooms with distorted walls and windows which appear (under fixed conditions) to be identical to rectangular rooms with rectangular windows.

In 1959, R. Penrose and independently J. Terrell showed that we won't perceive rapidly moving objects as Lorentz contracted but as rotated. Form V. Weisskopf's Physics Today review:

James Terrell (...) does away with an old prejudice held by practically all of us. We all believed that, according to special relativity, an object in motion appears to be contracted in the direction of motion by a factor $[1-(v/c)^2]^{1/2}$. A passenger in a fast space ship, looking out of the window, so it seemed to us, would see spherical objects contracted to ellipsoids. This is definitely not so according to Terrell's considerations, which for the special case of a sphere were also carried out by R. Penrose. The reason is quite simple. When we see or photograph an object, we record light quanta emitted by the object when they arrive simultaneously at the retina or at the photographic film. This implies that these light quanta have not been emitted simultaneously by all points of the object. (...) In special relativity, this distortion has the remarkable effect of canceling the Lorentz contraction so that objects appear undistorted but only rotated. This is exactly true only for objects which subtend a small solid angle.

Rudolf K. Luneburg, author of Mathematical Theory of Optics (1944), also published the book Mathematical analysis of binocular vision (Princeton UP, 1947) in which he argued that the "psychological space of binocular vision" carries a hyperbolic metric. From the Bull. AMS review:

Some of the topics which are treated in considerable detail are the horopter problem (geodesic lines), the alley problem, and rigid transformations of the hyperbolic visual space. (...) Using the hyperbolic metric, the author calculates the shape of distorted rooms which are congruent to rectangular rooms, that is, rooms with distorted walls and windows which appear (under fixed conditions) to be identical to rectangular rooms with rectangular windows.

Rudolf K. Luneburg, author of Mathematical Theory of Optics (1944), also published the book Mathematical analysis of binocular vision (Princeton UP, 1947) in which he argued that the "psychological space of binocular vision" carries a hyperbolic metric. From the Bull. AMS review:

Some of the topics which are treated in considerable detail are the horopter problem (geodesic lines), the alley problem, and rigid transformations of the hyperbolic visual space. (...) Using the hyperbolic metric, the author calculates the shape of distorted rooms which are congruent to rectangular rooms, that is, rooms with distorted walls and windows which appear (under fixed conditions) to be identical to rectangular rooms with rectangular windows.

In 1959, R. Penrose and independently J. Terrell showed that we won't perceive rapidly moving objects as Lorentz contracted but as rotated. Form V. Weisskopf's Physics Today review:

James Terrell (...) does away with an old prejudice held by practically all of us. We all believed that, according to special relativity, an object in motion appears to be contracted in the direction of motion by a factor $[1-(v/c)^2]^{1/2}$. A passenger in a fast space ship, looking out of the window, so it seemed to us, would see spherical objects contracted to ellipsoids. This is definitely not so according to Terrell's considerations, which for the special case of a sphere were also carried out by R. Penrose. The reason is quite simple. When we see or photograph an object, we record light quanta emitted by the object when they arrive simultaneously at the retina or at the photographic film. This implies that these light quanta have not been emitted simultaneously by all points of the object. (...) In special relativity, this distortion has the remarkable effect of canceling the Lorentz contraction so that objects appear undistorted but only rotated. This is exactly true only for objects which subtend a small solid angle.