The Shape of Change

[James Lemon]

 

[Jerome Marot]

I think that you have overcorrected the perspective a little bit. That gives the impression that the buildings are falling toward us. Is that what you wanted?

 

[James Lemon]

I think that you have overcorrected the perspective a little bit. That gives the impression that the buildings are falling toward us. Is that what you wanted?

Not sure what you are looking at ?

 

[Jerome Marot]

Vertical lines on the left building and on the right-hand side of the tower should be parallel to the sides of the frame. When one takes a picture with the camera looking up, they will go to the inside of the frame when looking up, as the sides of the letter A. Here they have been corrected, perhaps automatically, to go to the outside when following them up, as the sides of the letter V.

 

[James Lemon]

Vertical lines on the left building and on the right-hand side of the tower should be parallel to the sides of the frame. When one takes a picture with the camera looking up, they will go to the inside of the frame when looking up, as the sides of the letter A. Here they have been corrected, perhaps automatically, to go to the outside when following them up, as the sides of the letter V.

The vertical lines are straight . The slight sensation of tilt is more a consequence of visual psychology than an actual perspective problem.
What you're describing is a classic architectural perception effect.

The tower has several features that can create the impression of leaning even when it is actually vertical:

• The curved glass section draws the eye upward.
• The projecting balconies create repeating horizontal elements that are not perfectly symmetrical.
• The upper setbacks make the building appear narrower toward the top.
• The tower is surrounded by buildings with very strong rectangular 90° geometry.
• The beige low-rise in the foreground establishes a very stable visual reference frame.

When you place a tower with:

• curves,
• setbacks,
• projecting balconies,

next to a building that is mostly:

• straight lines,
• right angles,
• flat facades,

your brain starts comparing them. The more rectilinear building feels "true vertical," and the tower can appear to drift slightly even when it doesn't.

 

[James Lemon]

Vertical lines on the left building and on the right-hand side of the tower should be parallel to the sides of the frame. When one takes a picture with the camera looking up, they will go to the inside of the frame when looking up, as the sides of the letter A. Here they have been corrected, perhaps automatically, to go to the outside when following them up, as the sides of the letter V.

The left-hand building isn't a simple rectangular box. It has:

• multiple facade planes
• stepped setbacks
• different materials
• projecting elements
• varying window alignments
• architectural layers added at different times

So there is no single obvious vertical reference line running from bottom to top and those cues don't all agree with each other.

 

[Doug Kerr]

Best regards,

Doug

 

[Doug Kerr]

James, Jérôme,

There is in fact a very slight convergence upward on the image of what are certainly vertical lines on the building to the right.

But I suspect that the "sense" we may have that the vertical lines on the buildings significantly converge on the image is in fact an illusion, probably caused by the fact that we are used to seeing such lines coverage in our human sight, and we subconsciously "expect" that.

Best regards,

Doug

 

[Ted Cousins]

Trying to understand the context ...

was this image shot with the camera at eye-level and focused at the same level across the street, thereby making all verticals vertical?

Or, was it shot conventionally, focused somewhere up the building(s) and "corrected" in post?

I assume no tilt/shifting involved.

... sorry James, I didn't get the title.

rgds,

Ted

 

[Jerome Marot]

The vertical lines are straight . The slight sensation of tilt is more a consequence of visual psychology than an actual perspective problem.

Possibly. However, using rectangles as Doug showed, I still find that the vertical structures on the building are disposed as the sides of the letter "V".

 

[James Lemon]

Trying to understand the context ...

was this image shot with the camera at eye-level and focused at the same level across the street, thereby making all verticals vertical?

Or, was it shot conventionally, focused somewhere up the building(s) and "corrected" in post?

I assume no tilt/shifting involved.

... sorry James, I didn't get the title.

rgds,

Ted

The title "The Shape of Change" refers to the older commercial buildings in the foreground and the newer residential tower behind them. To me, the image shows how different eras of development shape the same street over time.

Regards,

James

 

[James Lemon]

Possibly. However, using rectangles as Doug showed, I still find that the vertical structures on the building are disposed as the sides of the letter "V".

I think what you're seeing may be the natural two-point perspective of the streetscape. The building is being viewed from a corner rather than straight on, so the geometry naturally opens outward toward the viewer. It's a perspective commonly seen in traditional cityscape paintings and architectural drawings.

 

[James Lemon]

There is in fact a very slight convergence upward on the image of what are certainly vertical lines on the building to the right.

But I suspect that the "sense" we may have that the vertical lines on the buildings significantly converge on the image is in fact an illusion, probably caused by the fact that we are used to seeing such lines coverage in our human sight, and we subconsciously "expect" that.

I think another factor is that we're looking at two very different architectural styles separated by many years of development.

The older street-front buildings are largely rectilinear, with strong horizontal and vertical elements, while the newer tower incorporates curved balconies, setbacks, and projecting forms. Because the two architectural languages are so different, the eye naturally compares them and may interpret the geometry differently than it actually is.

Part of what interested me about the scene was precisely that contrast between two eras of city building sharing the same block. That relationship is what inspired the title, The Shape of Change.

 

[Jerome Marot]

The older street-front buildings are largely rectilinear, with strong horizontal and vertical elements, while the newer tower incorporates curved balconies, setbacks, and projecting forms.

Now I understand the title. I don't think curved balconies in that style are used in Europe. I don't remember having seen any.

 

[Ted Cousins]

The title "The Shape of Change" refers to the older commercial buildings in the foreground and the newer residential tower behind them. To me, the image shows how different eras of development shape the same street over time.

Regards,

James

Understood - thanks!

Was the lens level or tilted up and fixed in post?

 

[James Lemon]

Understood - thanks!

Was the lens level or tilted up and fixed in post?

The lens was at level .

 

[Doug Kerr]

We of course need to recognize the possibility that the very slight divergence in the image of presumably-vertical lines in the scene is an artifact of distortion in the lens—its departure from perfect "rectilinearity".

I think the important issue here is that an image in which vertical lines of the scene appear essentially parallel my seem to us to have those lines "diverging" because we are used to looking at such a scene with our line of sight elevated from the horizontal.

Best regards,

Doug

 

[James Lemon]

We of course need to recognize the possibility that the very slight divergence in the image of presumably-vertical lines in the scene is an artifact of distortion in the lens—its departure from perfect "rectilinearity".

I think the important issue here is that an image in which vertical lines of the scene appear essentially parallel my seem to us to have those lines "diverging" because we are used to looking at such a scene with our line of sight elevated from the horizontal.

Best regards,

Doug

That's certainly a possibility. Lens characteristics can contribute small deviations from perfect rectilinearity, although in practical terms I suspect the effect here is quite minor.

What I find interesting is how much our perception influences the way we read architectural images. We are accustomed to viewing buildings with our line of sight elevated above the horizontal, so an image where the camera has been kept relatively level and the verticals remain essentially parallel can sometimes appear unusual, even when the geometry is quite accurate.

 

[Jerome Marot]

We of course need to recognize the possibility that the very slight divergence in the image of presumably-vertical lines in the scene is an artifact of distortion in the lens—its departure from perfect "rectilinearity".

The picture was taken with a Nikon Z8, which corrects such distortion automatically. Elements which are usually straight in a building are straight on the picture. It cannot be lens distortion.

Then, you may look at the windows on the left-hand side. They have white bars separating glass panels. You can draw a straight line between these bars on the 3 floors, thereby proving that the windows are aligned between floors. Yet that line is tilted.

There is no doubt in my mind that we see the effect of an overshoot of perspective correction. It is not an illusion due to building features, it is not lens distortion and it cannot be because the photographer was levitating above the ground. I found that puzzling and I asked whether it was intentional. It could have been. James made it clear it was not his intention.

Now back to the camera: Nikon Z8. That particular camera can correct perspective in camera. I don't know whether it has a tilt sensor or only corrects on command. Maybe there is something afoul with the camera or processing software: badly calibrated tilt sensor or correction left active when not needed, for example.

 

[Doug Kerr]

Hi, Jérôme,

The picture was taken with a Nikon Z8, which corrects such distortion automatically.

Ah, I didn't know that.

Thanks.

Best regards,

Doug