Jump to content

Was Patty's Arm Length "inhuman"?


Recommended Posts

This post was originally intended as my opening statement to a formal debate with Kitakaze regarding whether the length of Patty's arms can be considered "inhuman". I submitted it twice to the "PGF on Trial" moderated forum but they must have got lost in cyberspace so I decided to start a new topic with it instead.

I know, I know, as if we need another PGF thread. But the reason I'm starting this one is because I thought some members might be interested in the methods used to measure objects in photographic images. We can debate over the results, but not the methodology behind the science of photo-metrology.

Forensic photo-metrology uses "close range" photogrammetry to measure people in pictures. It is the science of measuring 3D objects projected onto a 2D photographic image. A 2D picture has no depth of field so the distances to objects gets lost. For example, a photograph taken from a plane aimed down at the earth will experience distortion from 4 sources: aberration from the lens, the camera not being square to the ground, the ground elevation and the curvature of the earth. Correcting for these combined distortions is called orthorectification. To orthorectify an image is to flatten it out so that all points on the surface are equidistant from each other. Then we can measure linear distances between objects. All mapping is done this way. In Patty's case, we are best to use frames that don't require rectification, because it's a complicated process that requires restoring the 3rd dimension, which isn't always doable.

To be clear, this topic is not intended to prove that Patty was a sasquatch. Likewise, no one can prove that Patty was Bob H in a suit from the PGF images alone. However, it is possible to exclude Bob H by measuring the images if they exceed his dimensions and they are not an illusion caused by the suit or the filming process. For Bob H to be in the suit, his dimensions must match to within an acceptable error tolerance.

There is only 1 way to measure a digital image of a scanned photograph and that is with photogrammetry. Peers can then refute or refine the measurements and maybe we can come to a consensus. But any graphic offered in rebuttal must adhere to the same methods. All the overlaid graphics submitted thus far do not measure body dimensions, including CGI physics engines that render posers or skeletons over images of Patty. Simply put, if a Poser matched Patty exactly, then it is possible that Patty's arms are 30% foreshortened and the Poser's arms are 20% foreshortened. In other words, even if their images align, their arm lengths might not. This is why CGI is not an appropriate measuring tool. CGI is missing the 3rd dimension and therefore cannot measure foreshortening. Matching the images up on a 2D surface does not rectify the image.

We must measure body parts that are unforeshortened or exceed a maximum unforeshortened length. Then if Patty's arms exceed a maximum length we can conclude that they must be longer than an average human's (control). Therefore, our goal is to test Patty's arms against human maximums and the question is then do Patty's arms exceed a human maximum relative to her height?

To begin we must assume Patty was a human in a suit. Then we can test whether she conforms to average human dimensions. For example, we should be able to determine the position of Patty's shoulder joint by fitting Bob H into the suit and noting where his shoulder joint aligns. If we assume it was Bob H in the suit then this effectively bypasses any "shoulder pad" claims. Also, we will bypass the shoulder joint entirely by measuring eye to elbow and eye to wrist distances. Note that these distances can be foreshortened to zero (if one points directly at the camera), but they cannot exceed their unforeshortened maximums. There is no such thing as forelengthening.

We also have to convert Patty's arm length into physical units, such as inches, then compare that against the control. The physical ruler will be Patty's standing height in pixels, which equals X pixels/inch. Then we can measure the distance from eyes to elbow, & wrist in pixels and convert to inches.

My first assumption is that Bob Heironimus can represent the "control" as an averagely proportioned human. I will also be using my own body dimensions to determine unforeshortened maximums for a 6' 3" human male. My arm span to height ratio is almost exactly 1, which is a human signature. I recommend all debaters measure their own body dimensions to confirm mine are average. You can be any height to establish relative body dimensions. As long as you are an averagely proportioned human.

Before I begin outlining the methods for measuring whether or not Patty's arms are "inhumanely" proportioned relative to her image height, we must first agree that the definition of "inhuman" is "not typically human", not "impossible for a human". I can only tell you what length Patty's arms are relative to her body, but I can't tell you that no human could possibly have arms her length. The converse contention is that Patty's arms are "humanly" proportioned. Let the measurements demonstrate whether this is true.

We must first establish Patty's arm length relative to her standing height. It turns out that any human's standing height can be estimated by adding the lengths of the vectors from the bottom of the foot to the knee, knee to the hip joint, hip joint to the top of the head. All these vectors must not be foreshortened significantly and parallel to the camera plane. This is confirmed by the following graphic.

tube_6ft2in_1225in_shoe.gif

Matt Crowley's runners were exactly 12.25" long, which makes the total length of his height vectors 6' 2". This was his standing height exactly. This provides an accurate estimate so it should be applicable for someone inside the Patty suit as well.

352StandingHgt.png

Patty's standing height was 8% taller than her walking height for frame 352. Draw a vertical line from the top of Patty's head (inside the suit approx 6" above the eyes) to the ground, level with the bottom of her left foot to confirm this. This is typically human for the lowest point in our walking gait. But was Patty at the lowest point in her gait? No. This is closer to the high point in her gait. Her lowest point exceeds 12%. The formula is 100 - Walking Height/Standing height * 100.

The only contestable issue is whether I placed the bottom of Patty's foot accurately. Otherwise, cite a better estimate but be warned that I extrapolated it to match Bob H's leg proportions. We are assuming it was him in the suit, after all. So if Patty's tibia (shin bone) is actually shorter than Bob's (which I suspect it is) then their height vectors won't match exactly and Patty would have to be scaled up to equal Bob's height. This would make Patty's arms even longer relative to Bob's. For this reason I'm confident the skeptics, at least, will accept my placement of the bottom of Patty's right foot. Else Patty would have to be scaled up even more.

Now that we have estimated Patty's image height we can use it like a 6' 2" ruler to measure her other body parts including the distance from her eye to wrist. We can also scale Patty to match Bob's height so their arm lengths can be compared. How about shoulder pads? Do shoulder pads extend the apparent length of the humerus (upper arm bone)? No, it's really not an issue because we can compare the height of the shoulder pads relative to the eyes, which are fixed, and it turns out if Patty had shoulder pads, they didn't extend the length of her arm appreciably. Same height from the shoulders as Bob H's shoulders. So there is actually no evidence that Patty had shoulder pads at all. And even if she did, they were understated. At any rate we will be measuring from the eye holes, which Bob H must have been looking through to navigate successfully. At least with his one eye. Eye to elbow, eye to wrist. That's the key.

Based on the following image, here are the numbers:

My unforeshortened maximums

------------------------

My eye to wrist = 36"

My eye to finger tip = 44"

My eye to elbow 25.5"

My elbow to wrist 10.5"

Based on the image above:

Conversion of pixels to inches = Physical standing height (74") / image standing height (489 pixels) = .1513

Patty foreshortened to an unknown degree

----------------------------------

Patty's eye to wrist = 267 pixels * .1513 = 40.4"

Patty's eye to finger tip = 321 pixels * .1513 = 46.6" (coarse estimate, fingers not visible)

Patty's eye to elbow = 181 pixels * .1513 = 27.3"

Patty's elbow to wrist = 88 pixels * .1513 = 13.3"

There you have it. All my unforeshortened measurements were exceeded by Patty's foreshortened estimates. And her arms were surely foreshortened to some degree, yet she beat my measurements just the same. What about Bob? The graphic below demonstrates the exact same differences between Patty and Bob. Bob doesn't match Patty's arms, yet according to his image his eye to wrist distance equals 35.5". Measure it against his height for yourself. This is almost my unforeshortened maximum of 36". Therefore, Bob's eye to wrist distance for the image below was almost maximal. There can be no excuses why his arm shouldn't match Patty's. The only defence is to claim that Patty's image was overscaled, but you are going to have to demonstrate that was the case. Good luck with that.

P_BH_elbow.gif

Note the estimated position of Patty's elbow above. It is a good 2 inches past Bob H's maximum eye to elbow distance. For all we know Patty's arm was foreshortened as well. Two inches might not seem like a lot, but it is significant to the "inhuman" claim. And that's only 1/2 the arm length. It probably accumulates to 5" longer to the finger tips than the control if Patty's fingers were visible in frame 352. How long must the arms be to be considered inhuman anyway? We never did establish the criteria for "inhuman". You be the judge. At any rate, it doesn't bode well for Bob H since the arms do appear to be disproportionate relative to his "average human" height.

BobPattyArmlength.png

The composite above is another comparison of Bob H to Patty. In this case I have attempted to use Bob H's eye to shoulder height to estimate where his shoulder was inside the suit. Doesn't look like the shoulder pads added much length to the arm. Therefore, they shouldn't skew the length of the humerus appreciably. But it doesn't matter because I'm using Bob's shoulder position anyways. I even used Bob's height to estimate the foreshortening of his arms and gave him some extra arm length to compensate. I gave him the benefit of the doubt at every turn. But once again he fell short. So did Bob H wear hand extensions? Does the radius (lower arm bone) look disproportionate? No. I think it is clear where the wrist is located from the following graphic:

ch.gif

Where is the elbow? Note the articulation point in the following graphic. The arm appears evenly proportioned to me. No extended humerus or radius. If anything the radius appears short or foreshortened to some degree.

elbowbend.gif

In summary, to determine the length of the arms relative to the body we need to apply metrology, not CGI. We must also test frames from the PGF that show minimal foreshortening, such as frame 352. We must then estimate Patty's standing height from head to toe. Then we assume that Patty's height was 6' 2 " and use her body as a ruler to measure the length of her arms. Also the distance from her eye to elbow and eye to wrist. Then compare those values against unforeshortened maximums from our control, Bob H (and myself).

I think I can predict how the skeptics will respond, but please stick to measuring the "arm length" using the method I outlined. Feel free to debate/revise the numbers with rationale, but let's measure the images using standard methods, ok? I realize that Kit claims to have found the suit (which I doubt), but IMO, a substantial stretch is required to put Bob H in it. :)

GF

Edited by Gigantofootecus
Link to post
Share on other sites
Guest BFSleuth

I think the common term for the span of hands (fingertip to fingertip) vs height is known as the "ape index" or "ape factor":

http://en.wikipedia.org/wiki/Ape_index

As noted by Giganto, the human standard is 1.0 (span = height).

A positive ape index is an advantage for many sports, such as swimming, boxing, or rock climbing. I tried to find a distribution of this factor for the human population to determine that which is "inhuman", and it is difficult to find anything worthwhile. Perhaps another forum member might be able to find that. The closest I could find is here:

http://www.ape-index.com/

Link to post
Share on other sites
PBeaton

Gigantofootecus,

Nicely done ! Your last comparison of Bob to Patty I find interestin', as she from the side appears near as wide as Bob viewed from the front, which would suggest her arms would also be further apart than Bob's. Don't mean to derail, but are you able to figure if that is correct ? Myself, no good with the measurin' stuff, it's all just observational on my part.

:drinks:

Pat...

Link to post
Share on other sites
Guest BFSleuth

Arm span will increase with shoulder width.

Another way to compare BH to Patty would be to compare the length of a single arm to the width between shoulder joints. I believe you will find that Patty's width will easily exceed BH's width. If this is established we can reliably rule out BH as Patty, and possibly rule out most of the human population.

Link to post
Share on other sites
Guest BFSleuth

Uh oh... now we may have tipped off the suit sleuths to the need to now add shoulder socket extenders to the suit... in addition to the upper arm extenders... and the lower arm extenders... and...

Link to post
Share on other sites
Guest RayG

Giganto, kudos on your presentation. Though I don't think it proves we're looking at a real sasquatch, I gave you a +1 for a lengthy and comprehensive post that was unburdened by distracting colors.

BFSleuth, maybe at the same time someone can explain John Green's measuring method*. It just doesn't make sense.

He says "My measurements of the film, made 36 years ago, gave the creature arms that were 30 inches from the shoulder to the wrist..." (Pay attention class, this is important. Patty's arm measures 30")

He adds...

"My own measurements are about 24 inches from shoulder to wrist..." (John Green's arm measures 24")

30" - 24" = 6" <-- that's math

So Patty's arms are 6" longer. <-- kind of important too

Here's where it gets interesting...

"...it is simple to calculate how many inches must be added to the arms of a man of known size in order to make his arms long enough to fit the supposed suit. In my own case the answer turns out to be about 10 inches."

Whaaaaa??? He's losing me. I thought his arms were only 6" shorter?? Apparently it's not so simple anymore. Is this new math? Seriously, why is he adding 10" when the difference is only 6"?

He continues...

"But in order for the arms to bend at the elbow, which they plainly do in the movie, all of that extra length has to be added to the lower arm.

The result, in my case, is about 12 inches of arm above the elbow and 29 inches below it—almost as much of a monstrosity as Edward Scissorhands."

Ok, he's lost me altogether. Adding 10" to his own arm [24" + 10" = 34"], regardless of the location, is still 4" too much. What on earth does that have to do with him having an arm forty one inches long [12" + 29" = 41"], which is 11" longer than the length he claims for Patty's arm and 17" longer than his own?

24" + 17" ≠ 30"

Math is pretty straightforward. Numbers either add up or they don't, and in John's case they don't. I know Roger Knights took a stab at explaining John's math once upon a time, but it seemed to be based on assumptions not math, if I recall correctly. Anyone else care to have a go?

RayG

* http://www.bfro.net/...lenge/green.asp

Link to post
Share on other sites
Guest BFSleuth

I would rather stick with the OP and keep Green's analysis out of it. Giganto seems to have a handle on the math. If you want to start a separate thread to rip apart Mr. Green fire away...

... after all, this was Giganto's post that was apparently intended for debate. Better to allow this thread to focus on the methods that Giganto has brought forward.

Link to post
Share on other sites
Guest RayG

I have no trouble with folks focusing on Giganto's initial post, I was just pointing out that claims of 'inhuman' Patty proportions aren't always based on valid math.

RayG

Link to post
Share on other sites

I think the mods should consider removing the reference to Green's math, it really has no place here, and will just allow for derailment.

I think shoulder width, and body thickness,verses arms length etc, is very interesting. Something I had not thought about before. The shear mass of space someone would have to occupy within the suite, and still make it function.

Link to post
Share on other sites

I would disagree John. The post is about using mathematical models to estimate the length of objects is a photo. It looks like Giganto's math "adds up" and comparing it to past, faulty methods does not diminish his method. I will anxiously await a well reasoned rebuttal from skeptics.

Link to post
Share on other sites
Guest MikeG

Guys, BFSleuth has given the mods' response. Please stick to discussing Gigantofooticus' maths and methods, and please don't discuss other people's here, nor, indeed, moderation itself.

Feel free to start other threads to discuss other people's version of the same idea. Please feel free to bring issues with moderation to our attention by PM.

Mike

Link to post
Share on other sites
  • gigantor unlocked this topic
×
×
  • Create New...