De-bouncing the Zapruder film
Herbert Blenner
You may be asking what is de-bouncing? In the technical jargon they de-bounce
switches so why not de-bounce a jumpy motion picture. Well, last August with
the heat hundred around 110, staying in my air-conditioned apartment and
working with the still frames of the Z-film seemed like a good idea.
Projecting a series of still frames on a fixed viewport shows camera jingle and
perhaps frame misalignment from digital conversion. Regardless of cause, the
bouncing obscures many details recorded on the film.
Using a moving viewport corrects both sources of bounce. However, the position
of the viewport must be adjusted for specific details on each individual frame.
I focused my initial effort on the movements of Toni Foster. After finding the
correct positions for the viewport on frame Z-309 to Z-317, the improvement in
intelligibility encouraged me. The moving viewport compensated for the apparent
leftward movement of Toni Foster. This compensation revealed otherwise
imperceptible details of Toni Foster's gait. I decided to apply the de-bounce
technique to the head shot.
Initially, I picked a detail on the limo as a reference point. After adjusting
each individual viewport to kept the reference point stationary, I found the
changing angle between the camera and people in the limo limited the
improvement in intelligibility. As a compromise, I used several points on the
limo and adjusted individual viewpoints for smooth rotation of the limo. This
approach worked well and clearly showed movement of the debris cloud and
separated the forward from the reverse body movements.
At this point of refinement, I recognized frame rotation as a new limiting
factor. Individual frames differ by their rotation about an axis is
perpendicular to the field of view. In principle, software can rotate images
about this axis and correct the problem. However, these angular errors are
minor compared with frame rotation about an axis parallel to the field of view.
Here the angular errors changed the recorded information, making software
correction impossible. As a result of this complication and cooler weather, I
abandoned any angular corrections.
My next project for the snowbound days of winter will be analysis of the
movements of Governor Connally between frames 250 and 290. Preliminary
examination of these frames show a tantalizing discontinuity in the forward
motion of Connally around Z-271.
Martin Shackelford
Martin
Herbert Blenner
>From: Martin Shackelford msh...@concentric.net
>Date: 10/10/02 7:52 PM Eastern Daylight Time
>Message-id: <3DA61339...@concentric.net>
>
>What you describe was done by Robert Groden in the 1970s.
>
Do you mean it took them ten years to figure out what was done in the fifties?
I think you give them too much credit. They
had help.
vbg,
Herbert
>Martin
>
>
>
>
>
>
GordonD
Herbert,
I've been trying to get martin to review a few SMPTE [Society of
Motion Picture and Television Engineers] journals [at most University
libraries] preferrably from the 30's -40's era, regarding, just WHAT
techniques were available for the artform called optical film
printing.
The fact is; the artform, the equipment, the operators/technicians
were there in the earlt 60's to do a job on the Zapruder film.
See Raymond Fieldings [Prof at Univ. of Iowa] The Technique of Special
Effects Cinematography - Lib of Cong. Catalog Card 64-8116 published
in 1965, the first book published in English in the USA covering the
techniques...of the art
Evidently the university Martin uses have none of these resources, or
it's just to late for him to look at unimpeachable resources regarding
the REAL possibility of alteration to the Z-film. Not to mention
techniques.
David Healy
> Herbert
>
> >Martin
> >
> >
> >
> >
> >
> >
Martin Shackelford
copies of the film, goes back earlier than the '50s. Disney used it in
their early animated features.
Martin
Martin Shackelford
Have you read the Zavada Report yet?
Martin
Herbert Blenner
>From: Martin Shackelford msh...@concentric.net
>Date: 10/12/02 3:53 AM Eastern Daylight Time
>Message-id: <3DA7D554...@concentric.net>
>
>You keep missing the point, David.
>Have you read the Zavada Report yet?
>
Of course David has read the Zavada Report. It is required reading for any
public relations course.
Herbert
Martin Shackelford
Your last couple of posts could use translations. Either you're
just being sarcastic, or you're mumbling total nonsense--or both.
Martin
Herbert Blenner
>Message-id: <3DA91159...@concentric.net>
>
>Herbert,
>
> Your last couple of posts could use translations. Either you're
>just being sarcastic, or you're mumbling total nonsense--or both.
>
You keep missing the point, Martin.
Have you read the Sarcasm Manuel yet?
Herbert
>Martin
>
>
>
>
>
>
GordonD
> >Subject: Re: De-bouncing the Zapruder film
> >From: Martin Shackelford msh...@concentric.net
> >Date: 10/12/02 3:53 AM Eastern Daylight Time
> >Message-id: <3DA7D554...@concentric.net>
> >
> >You keep missing the point, David.
> >Have you read the Zavada Report yet?
> >
>
Martin,
Yes -- I've read the Zavada report, on more than one occasion ...
Rumor has, I may be discussing aspcts of the report with Roland Zavada
next May sometime.
Thank you Herbert , it is THAT, indeed:-)
David Healy
Martin Shackelford
It's unfortunate that you have already reached your conclusion before
doing so, however.
GordonD
What's unfortunate is, 'painting me into a corner' by assuming I'm
convinced the Zapruder film is altered.
My position regarding this issue is clear today as it was years ago,
the possibility of alreration can NOT be denied -- the optical film
printing equipment, personnel and the know how, not to mention the
TIME required, were INDEED available in 1963/1964. Not to mention 20
years BEFORE that!
Which of course flies in the face of positions taken by many on this
and other forums.
Roland Zavada, despite his film matters expertise will provide no
concrete proof either way regarding this ' it IS - it ISN'T ' issue.
Considering reasons [real or imagined] which side of the fence
benifited from alledged alteration -- well, it's pretty obvious to me
-- but then again -- NO proof.
Quite frankly, I don't believe the *camera original* Zapruder film
exists, for that matter may have never left Dallas.
Martin Shackelford <msh...@concentric.net> wrote in message news:<3DAA5AD3...@concentric.net>...
AnthonyMarsh
The method of alteration you are considering about is physically
impossible to create the Zapruder film we see. Read my article about the
sprocket hole images.
http://www.boston.quik.com/amarsh/zapruder.htm
--
Anthony Marsh
The Puzzle Palace http://www.boston.quik.com/amarsh
Martin Shackelford
Martin
GordonD
And ah, just what method is it that I'm considering??? Physically
impossible? roflmfao ... You know little of optical film printing ---
Physics isn't reality here!
What ever you see in the Zapruder film can be recreated, including the
sprocket hole area...
AnthonyMarsh
I know a lot about how the ghost images were created, which makes your
idea impossible. Do you happen to know Robert Groden, who is the one who
actually did the rotoscoping? I do. Why not visit his web site and ask
him directly.
Did you read my article or not?
> What ever you see in the Zapruder film can be recreated, including the
> sprocket hole area...
Herbert Blenner
>Date: 10/15/02 7:35 PM Eastern Daylight Time
>Message-id: <3DACA64D...@quik.com>
Anthony Marsh wrote:
Yes, Anthony, I have read your article. You discuss how the cyclic operation of
the camera produces ghost and sprocket area images. Technically, they describe
these images as the convolution of the present image with portions of other
images.
Typically, they describe any cyclic machine by as a relationship between the
present output and a weighted sum of past, present and future inputs. In the
case of a motion picture camera, the present output corresponds to the recorded
frame with ghost and sprocket images. The past, present and future input are
the ghost free images. Finally the weight functions associated with the past
and future inputs represent the specifics of the camera mechanism.
One camera may have a vanishing weight function for the past image while
another has a null weight function for the future image. Still, a third camera
could have two nonzero weight functions. These weight functions, however,
uniquely describe ghost and sprocket image production.
The ideas, presented above represent what I learned as an undergraduate
engineering student during the mid-sixties. However, Dr. Tou wrote the course
book during the late fifties. So much for your claim that during the sixties
nobody understood ghost image production. During the sixties engineers were
designing machines to solve the deconvolution problem.
When dealing with a camera, deconvolution finds ghost-free images given the
ghosted images and the weight functions of the camera. Generally, this
calculation is difficult. However, the multilayer construction of the color
film permits an accurate optical solution. They individually copy each primary
color to a separate film then use these films as ghost filters. When they
project images from a ghosted film through the ghost filters, a third film
records nearly ghost free images.
Contrary to the opnion of your cited experts, ghost production, removal, and
reproduction was not only possible during the early sixties, it was part of the
curriculum at our colleges.
Herbert
GordonD
Why should I speak to Robert Groden about editing/compositing? I've
done enough film/video compositing over the past 35 years to know what
I'm talking about... I wish I could say the same for all the **
supposed ** film experts [not] around here.
Times are-a-chang'in -- looks like it's out with the old guard and in
with then new...
AnthonyMarsh
"Convolution"? What the hell are you talking about. The ghost images are
double exposures, just as hoaxers used to produce ghost images by
employing double exposures. Same concept. In the case of the Zapruder it
is an accidental effect of the shape of the plate opening and the fact
that the camera was on full telephoto.
> Typically, they describe any cyclic machine by as a relationship between the
> present output and a weighted sum of past, present and future inputs. In the
> case of a motion picture camera, the present output corresponds to the recorded
> frame with ghost and sprocket images. The past, present and future input are
> the ghost free images. Finally the weight functions associated with the past
> and future inputs represent the specifics of the camera mechanism.
>
You are spouting pure gibberish to see if I am impressed enough to think
that you are an expert. I am not and you are not.
> One camera may have a vanishing weight function for the past image while
> another has a null weight function for the future image. Still, a third camera
> could have two nonzero weight functions. These weight functions, however,
> uniquely describe ghost and sprocket image production.
>
What you are saying is nonsense. The ghost images are just double
exposures.
> The ideas, presented above represent what I learned as an undergraduate
> engineering student during the mid-sixties. However, Dr. Tou wrote the course
> book during the late fifties. So much for your claim that during the sixties
> nobody understood ghost image production. During the sixties engineers were
> designing machines to solve the deconvolution problem.
>
No one here ever made such a claim. And Zavada was a consultant to a TV
show about how hoaxers used to produce ghost pictures often using double
exposures.
> When dealing with a camera, deconvolution finds ghost-free images given the
> ghosted images and the weight functions of the camera. Generally, this
> calculation is difficult. However, the multilayer construction of the color
> film permits an accurate optical solution. They individually copy each primary
> color to a separate film then use these films as ghost filters. When they
> project images from a ghosted film through the ghost filters, a third film
> records nearly ghost free images.
>
Pure nonsense.
> Contrary to the opnion of your cited experts, ghost production, removal, and
> reproduction was not only possible during the early sixties, it was part of the
> curriculum at our colleges.
>
Producing ghost images by double exposure has been done since the advent
of photography.
Martin Shackelford
Martin