Digital photo stitching software program is the workhorse in the panorama-making process, and can vary from providing a completely automatic one-click stitching, to your more time-consuming manual process. This is certainly part 2 of your tutorial, which assumes all individual photos are actually properly captured (stage 1 below is done); for stage 1 and a review of the complete stitching process please visit part 1 with this tutorial on digital panoramas.
As a way to begin processing our combination of photos, we should select a proper software program. The most significant distinction between options is in the direction they elect to address the tradeoff between automation and adaptability. Generally, fully customized stitching software will always achieve higher quality than automated packages, but this may also result in being overly technical or time consuming.
This tutorial aims to further improve understanding of most software stitching concepts by maintaining the discussion as generic as is possible, however actual software features may talk about a treatment program called PTAssembler or PTGui (front-end for PanoTools or PTMender). PTAssembler incorporates a fully-automated one-click stitching option, together with providing for almost all possible custom stitching available choices in other programs.
During the time of this article, other notable programs include those who come packaged together with the camera, such as Canon PhotoStitch, or popular commercial packages including Autostitch, photo making software, Arc Soft Panorama Maker, Panorama Factory and PanaVue, and others.
STAGE 2: CONTROL POINTS & PHOTO ALIGNMENT
Panorama stitching software uses pairs of control points to specify parts of two camera photos that reference the identical reason for space. Pairs of control points can be manually selected by visual inspection, or these might be generated automatically using sophisticated matching algorithms (such as Autopano for PTAssembler). With most photographs, best results could only be achieved with manual control point selection (which is usually the most time-consuming stage from the software stitching process).
The example above shows a selection of four pairs of control points, for 2 photos in a panorama. The most effective control points are the types that are based upon highly rigid objects with sharp edges or fine detail, and so are spaced evenly and broadly across each overlap region (with 3-5 points for each overlap). Because of this basing control points on tree limbs, clouds or water is ill-advised except when absolutely necessary. It is actually because of this recommended to continually capture some land (or some other rigid objects) within the overlap region between all pairs of photographs, otherwise control point selection may prove difficult and inaccurate (for example for panoramas containing all sky or water).
The example below demonstrates a scenario the location where the only detailed, rigid section of each image is with the silhouette of land on the very bottom-thereby which makes it tough to space the control points evenly across each photo’s overlap region. Over these situations automated control point selection may prove better.
PTAssembler has a feature called “automatically micro-position control points,” which works by with your selection as an initial guess, then seeking to all adjacent pixels inside a specified distance (like 5 pixels) to ascertain if these are typically a greater match. When stitching difficult cloud scenes including that shown above, this effectively combines the advantages of manual control point selection with the ones from automated algorithms.
Another consideration is the way far awayfrom the digital camera each control point is physically located. For panoramas taken with no panoramic head, parallax error may become large in foreground objects, therefore better results may be accomplished by only basing these on distant objects. Any parallax error from the near foreground may not be visible if each one of these foreground elements are not contained within the overlap between photos.
STAGE 3: VANISHING POINT PERSPECTIVE
Most photo stitching software gives the ability to specify where the reference or vanishing reason for perspective is found, in addition to the form of image projection.
Careful choice of this vanishing point can help avoid converging vertical lines (which could otherwise run parallel), or a curved horizon. The vanishing point is usually where one would be directly facing when they were standing inside the panoramic scene. For architectural stitches, like the example below (120° crop from your rectilinear projection), this point is also clearly apparent by simply following lines into the distance which can be parallel to one’s brand of site.
Incorrect placement in the vanishing point causes lines laying from the planes perpendicular on the viewer’s line of site to converge (although these would otherwise appear to be parallel). This effect can even be observed by using a wide angle lens inside an architectural photo and pointing your camera significantly below or above the horizon- thereby giving the impression of buildings that happen to be leaning.
The vanishing point is additionally critical in very wide angle, cylindrical projection panoramas (like the 360 degree image shown below). It may well exhibit different looking distortion if misplaced, causing a curved horizon.
In the event the vanishing point were placed too high, the horizon curvature would be in the opposite direction. Sometimes it might be challenging to locate the specific horizon, as a result of presence of hills, mountains, trees or some other obstructions. For such difficult scenarios the location of the horizon could then be inferred by placing it at the height which minimizes any curvature.
Panorama stitching software also often provides the option to tilt the imaginary horizon. This is very useful when the photo containing the vanishing point had not been taken perfectly level. For this particular scenario, even if your vanishing point is put at the correct height, the horizon could be rendered as having an S-curve in the event the imaginary horizon fails to align with the actual horizon (from the individual photo).
When the panorama itself were taken level, then a straightest horizon will be the one which yields a stitched image whose vertical dimension will be the shortest (and is a technique sometimes used by stitching software).
STAGE 4: OPTIMIZING PHOTO POSITIONS
Once the control points, vanishing point perspective and image projection supply been chosen, the photo stitching software can then commence to distort and align each image to produce the very last stitched photograph. This can be the most computationally intensive step in the process. It operates by systematically searching through mixtures of yaw, pitch and roll in order to minimize the aggregate error between all pairs of control points. This method could also adjust lens distortion parameters, if unknown.
Keep in mind that the aforementioned photos are slightly distorted; this really is to emphasize that if the stitching software positions each image it adjusts for perspective, and therefore the amount of perspective distortion depends on that image’s location in accordance with the vanishing point.
The key quality metric to understand may be the average distance between control points. If this type of distance is big relative to paper size, then seams may be visible regardless of how well these are generally blended. One thing to check is whether any control points were mistakenly placed, and that they follow the other guidelines listed in stage 2. In case the average distance continues to be too large then this can be caused by improperly captured images, including parallax error from camera movement or otherwise utilizing a panoramic head.
STAGE 5: MANUALLY REDIRECTING & BLENDING SEAMS
Ideally one would like to put the photo seams along unimportant or natural break points throughout the scene. In the event the stitching software supports layered output one could perform this manually employing a mask in photoshop:
Without Blend Manual Blend Mask from Manual Blend
Note exactly how the above manual blend evens the skies and avoids visible jumps along geometrically prominent architectural lines, for example the crescent of pillars, foreground row of statues and distant white building.
Ensure that you blend the mask over large distances for smooth textures, for example the sky region above. For fine detail, blending over large distances can blur the picture if you have any misalignment between photos. It really is therefore best to blend fine details over short distances using seams which avoid any easily noticeable discontinuities (observe the “mask from manual blend” above to discover how the sky and buildings were blended).
On the flip side, manually blending seams could become extremely time consuming. Fortunately nikon photo software has an automated feature that may perform this simultaneously, as described within the next section.
STAGE 5: AUTOMATICALLY REDIRECTING & BLENDING SEAMS
One of the best strategies to blend seams within a stitched photograph is to use a method called “multi-resolution splines”, which may often rectify even poorly captured panoramas or mosaics. It operates by breaking each image up into several components, just like how an RGB photo might be separated into individual red, green and blue channels, except that in cases like this each component represents an alternative scale of image texture. Small-scale features (such as foliage or fine grass) have a superior spatial resolution, whereas larger scale features (like a clear sky gradient) are said to obtain low spatial resolutions.