If you found an old picture and wish to take a shot of the after-picture, it is best to first make a printout of the before-picture or copy it to your smartphone. Once on site, you then take a sample shot and compare it with the old photo, and then correct your position if necessary.
In order for the images to overlap well after subsequent alignment, it is important that the perspectives from which the two photos were taken are as close as possible to one another. The more the perspectives differ, the more distorted the images will be later. The transformation capacity using markers is not unlimited and depends on this first step.
Overlays as in the examples give clear spatial information, i. e. if the relative positions in this part of the picture are correct, the perspective is correct with respect to the whole picture. To be on the safe side, you can also look for another reference point.
This strategy may sound complicated at first, but with a little practice you will get an eye for these details over time. Needless to say, it is not possible to find overlays for all images, as seen in example 3:
This is obviously not an exhaustive guide for all motives. With time you may find your own strategy, which you can share in the forum!
It is advisable to take several pictures, especially if you are not sure about the position, and then to select the one with the best matching perspective later on the PC. It is best to do this systematically: For example, after you have taken a picture, go one step to the left, take the next one, go another step to the left, and retake, and so on.
Also consider the following general tips for taking pictures:
Our image transformation is designed to ensure that both pictures match pixel-perfectly. To do this, use the mouse to place at least 4 colored markers on conspicuous spots which are same both on the before and after picture, e.g. the edge of a house or the tip of a tower. Two things are to be considered in this process: * The markers should be distributed as evenly as possible over the image area. * Markers should be set both in the foreground and in the background with respect to the image depth. This example should clarify this:
In this case, markers were distributed over a large area, both in the foreground and in the background. At least four markers must be placed. More markers can improve the alignment (how many additional markers are useful depends on the image).
If in the result the pictures at the markers are not overlapped, although the markers have been placed at the matching places, there is a little trick which is sometimes helpful: You can drag the marker in the new picture in the direction where the old picture is still to be shifted (like a net you are dragging). The markers will then no longer be located at matching points, but the overlay will improve. The precondition, however, is that the perspectives are the same and one must not exaggerate, because otherwise the images will be excessively distorted.
If you have any questions or need help in taking or aligning your pictures, please do not hesitate to ask the Forum!
Author: Nicolai Wolpert
If you would like to rate other users’ pictures or upload your own pictures, you have to register.
Each user must pick a username, an email address and a password. Optionally, a display name, a location and a time zone may be chosen.
By default, the user name is listed as author of an image in the tabular view and in the single view. If the user has provided an optional display name, it will be used instead of the user name in the single view.
Take a printout of the before-photo with you and try to find the original spot by comparing the printout with the live camera view.
In general: portrait mode is less suitable, since during presentation it generates black stripes left and right of the photo. Pictures in landscape modes fit best on a typical 16:9 display, if they respect the same ratio. Therefore we have optimized the single view display for the presentation of a 16:9 photo next to the Google map. Other ratios like 4:3 are also possible, however they waste a little more space.
Both the before- and the after-photo are uploaded and may/must be supplemented with a category (buildings, interior, nature, people), a title, a description, a location and the two recording dates. Important: the user must have the authorization to publish both pictures.
If you don't know the exact date of a photography, you may guess the year and set it to "Exact time unknown". Data for month, day, minute and second may be set to "unknown". If provided, information for location and time may be extracted from the digital image EXIF data.
If the user has recorded the picture himself, he can choose the type of license. The most restrictive would be "All rights reserved", which means, the photo may be inspected, but not distributed without explicit approval of the author. A more liberal approach would be "CC BY-ND". This Creative Commons license allows the distribution of the photo, in a commercial and non commercial setting, as long as the author is clearly stated and no modifications are made.
If you use a picture provided by an archive, which comes already with a copyright license, it is your duty to keep that license attached to the image and propagate it further.
Our image transformation makes sure, that both photos fit together pixel by pixel. For this reason, please place with the mouse up to 4 colored markers on the before-photo and the after-photo at distinctive spots, that represent the same location in the scene, e.g. the corner of a building or the top of a tower. After aligning the two pictures with our software, you may decide whether you are happy with the result or whether you would like to start all over again with a new set of markers. Be sure, to place markers both in the foreground and in the background.
All image pairs can be filtered by the categories, by the time interval boundaries and by the time interval duration. Terms within title and description can be searched as well as for user names. The resulting hits will be displayed in tabular form and will be georeferenced on a Google map. The table can be sorted column by column. You may zoom in and out of the map.
By clicking on the head of a table column all entries are sorted with regard to that attribute.
You may move a slider with the mouse. You may rate the image with up to 5 stars. You may comment on the image by typing a few sentences in a text field below the last comment.
Each user may award up to 5 stars to each image pair, that is not authored by himself. The average of all the stars given to an image results in the average rating, provided, there are enough ratings.
The forum is a place for photographers to talk about their experiences and ideas in the context of before-after-photos: Where do I get historical pictures ? What is important during the shooting of a picture ? Which cameras are suitable ? What are fancy motives ?