is the fist plugin package of a brand new series of plugins for Final Cut Pro X and Motion 5. This package is a 100% redevelopment of our popular Fx-Script plugin package "The Repair-collection
". Final Cut Pro X and Motion 5 are not able to run Fx-Script plugins like "The Repair-collection", which is the reason why this plugin package was completely redeveloped from the ground up for Final Cut Pro X. As an additional advantage of the redevelopment you are now even able to use these plugins with Motion 5.Repair FX
includes these four plugins:
- Noise Reduction FX
Dead Pixel FX
Dirt Remover FX
Dropout Eliminator FX
This webpage describes how these plugins work and how to use them.
Noise Reduction FX
Get rid of noise and dust
Noise Reduction is a very powerful tool, but yet only has four controls.
How does it work?
Noise Reduction is different to all the others de-noise tools out there. Most other tools simply apply a frame by frame blur over the video. Even if those tools avoid to blur and edges, they will still just blur the video. If you apply a simple Sharpen filter then you will get all the noise back. Noise Reduction FX works different because it takes a completely different approach. Noise Reduction FX scans a certain timespan, compares all the frames of that timespan to each other and evaluates mathematically which part of the video is real content and which part is noise that needs to be eliminated. By going this way you can get much better results as you have ever seen with Final Cut Pro or Motion.
To make life simple Noise Reduction FX only carries 4 controls:
Noise (to eliminate)
Here you can set the maximum amount of noise you want Noise Reduction FX to eliminate. Note: Don't set this control too high. A too high setting might also eliminate content that should stay in your video.
To demonstrate how the Noise parameter works we will use a really bad video. A simple blur filter would have to give up on this one. In the background you can barely see some cable cars. To the left there is the tower from a crane.
Now we set Noise to 15. You can see that the noise is nearly gone, but the content of the video is still there. You can clearly see the cable-cars and the tower. Even the cables of the cable cars are now visible.
Let's set Noise to 50. What happens now? Noise Reduction now also tries to eliminate the "noise" it can see in the movement of the cable cars and (due to the moving camera) the moving crane tower. In this case a value of 50 would be a too high because it also eliminates the content of the video.
What we learn from this example is that you have to be careful with the Noise value. Too high values can destroy the footage, too low values have no effect. Values from 2 to 40 are good in most cases depending on the video and the amount of noise. Higher values reduce more noise, but also might eliminate content of the video. If the value is set too low, you will still see your noise. Suggestion: Play with this slider. Noise Reduction always shows you the final rendered video in the canvas. Begin with lower values and evaluate the best setting.
This slider tells Noise Reduction how much video should be taken into account for the detection of the noise. A longer timespan looks at more surrounding frames, but might also add a little motion-blur effect. Good values go from 10 to 40.
Noise Reduction works by taking surrounding frames into account for the evaluation of the existing noise. If the current scene of your video changes (i.e. due to a cut), it is not useful for Noise Reduction to use these frames for the evaluation. Therefore Noise Reduction calculates a value that describes the difference between the single frames. The higher the value, the more likely it is that the scene has changed. This is what Sensitivity is for. This slider determines a threshold and tells Noise Reduction the maximum allowed amount of the difference between the frames. You don't need to cut you footage into little scenes. Noise Reduction takes care of scene changes automatically and makes it possible to eliminate the noise of a whole film at once.
If you check the Calc and show checkbox, you will see this additional info in the canvas:
The black marker in the center marks the current frame. The white surrounding dots represent the surrounding frames. A white dot means that the frame is taken into account for the noise evaluation. A red dot means that the frame will not be used.
The Sensitivity value to the right is the calculated difference between the current frame and the next frame.
Here we can see some frames that are marked in red color. These frames will not be used for the noise evaluation, as they are too different from the current frame anyway.
Here we can see a typical scene change. The Sensitivity value for this frame to the next frame is 15. This means that the content of the video did not change a lot between those two frames. But there is a big change to some earlier frames. This is a typical display after a scene change.
Hint: Remember that noise will also show up in the Sensitivity value as it boosts up the value by the amount of the current noise. Even if you have a completely static scene, you will see that the Sensitivity value is not at zero. The value you see at that frame represents the actual noise and can be used to determine a good value for the Noise (to eliminate) slider.
Dead Pixel FX
The intelligent tool to repair dead pixels and spots
Dead Pixel is able to reconstruct damaged areas with a size of up to 16 x 16 pixels. Normally you will use this tool to repair footage that has been shot with a digital camera with dead pixels. Such footage is actually useless and in most cases you can't bring it back. If you have such footage, Dead Pixel will be a life saver.
At first you have to apply the filter to the footage. Open the controls of Dead Pixel and select if you use interlaced or progressive video with the "Source Type" popup menu. Then set the Mode popup menu to "Mark Dead Pixel".
Now mark the pixel that need to be repaired in the canvas. Use the control "Upper Left Corner" to set the marker in the canvas as close as possible to the damaged pixel (or area). A useful magnification window in the canvas shows a magnification of the area where the control Upper Left Corner is set.
Remember to set the canvas to at least 100%. If you use a smaller size, you won't see the whole video resolution and you probably wont be able to place the marker at the right spot.
Don't worry if you could not set the marker at the perfect spot right away. The Additional Offset sliders enable you to shift the marker perfectly on top of the desired spot pixel by pixel. The additional magnification window in the canvas also helps to mar the spot exactly.
Size of Dead Pixel
Here you can determine the actual size of the dead pixel (or damaged area). If you have set the marker at the upper left corner of the damaged area, you will see that the damage vanishes while you are adjusting the size of the marker.
The restoration progress can be monitored while adjusting the size of the marker.
If you have footage with a single dirty spot, you can also use Dead Pixel to repair it. Isolate the damaged frame with the blade-tool (press B) and make a cut at either side of the damaged frame. Apply Dead Pixel and go to the controls. Move the marker to the damaged area and adjust its size. You will see that even bigger damages can be repaired.
Dirt Remover FX
Remove black and white spots from your video automatically
Dirt Remover is made to eliminate dark and bright spots from your video and replace them with content taken from nearby frames. You can choose between the elimination of black, white or both types of spots at once.
In order to remove occasional dirt spots Dirt Remover has to look at nearby frames to find out if a certain spot is really a spot or a freckle or a moving object. If a spot has been identified, it will be replaced by the content from the last or the next frame or a combination of both. Therefore Dirt Remover creates a mask where all identified spots can be seen. You can preview the mask in the canvas.
There are three modes:
The canvas shows you the unchanged footage. This is very useful to see if you have used good settings that only took care of the dirt. In this case we can see a video with some dark spots within this frame.Mode: Removed Dirt (Mask)
Here we see the spots taken from the video. This will be used as a mask for the replacement with content from nearby frames.
The canvas shows the actual rendered footage. You need to switch to this setting when you want to render the image.
This popup is used to tell Dirt Remover which frames should be used to overlay onto the areas where spots have been detected.
The option Last and next frame 50/50 will add up the previous and the next frame. This option is the best way to be sure to remove a spot completely by original content. If i.e. the video changes its brightness during that time, the 50/50 option adds up the surrounding frames and so creates a frame that carries the exact brightness that got lost by the unwanted spot.
Use this popup to determine if you want to remove black or white spots or both at once.
Here you can fix the maximum size of the spots you want to have removed. Only spots smaller than the given size will go away. All other spots will stay. This value is most important. Try to imagine the following: You have a very fast moving video where the content changes rapidly. A hand appears for only three frames. In the first frame the hand is visible on the left side of the video, in the second frame in the center and in the third frame on the right side. Dirt Remover must now determine if the hand in those frames is really some content that should stay or a spot that needs to be removed. This is done by telling Dirt Remover the maximum size of a spot. Usually such a fast moving object is way bigger than a spot you want to get rid of. By setting the Dirt Size slider to the maximum size of the spots, you can ensure that only spots will be removed and such a fast moving hand will stay where it is.
In order to get a good grip on real spots, Dirt Remover needs to evaluate the content of the video. It is very easy for Dirt Remover to see where a spot appears if there is no motion in the video. But if the video has a very fast motion, it is more complicated to figure out if a specific object is a spot or just a fast moving freckle. This is why there is the Motion Tolerance control. The higher you set this value, the less will Dirt Remover take care of the motion of the objects in the video and it is possible that some objects will be removed. Make sure not to set this value too high. Values between 1 and 40 are good in most cases.
Dirt Brightness Range
Dirt does not always look the same. In addition each video might have a different black and white level. Use these two controls Low and High to tell Dirt Remover the brightness range of the video. Don't narrow the brightness range too much. Dirt Remover needs to have a chance to look at the whole video. A too narrow range is contra productive and should be avoided.
This additional control might help in some cases when you get borders around the repaired spots. With this control you can feather the mask in order to eliminate borders around the repaired areas.
Dropout Eliminator FX
Repair even the worst dropouts
Dropout Eliminator repairs damages like dropouts or burn holes. If your video contains big damages in a single frame or field, you can replace the lost content with the content of a nearby frame. As a requirement your video must have a similar frame close to the damaged frame. You might understand that there is not even a chance for this plugins to repair a big dropout if there is nothing that can be used to fill the damaged hole.
At first you have to determine if the damaged area is visible only in a single field or in a full frame. If you use interlaced footage it is most likely that the damage is only visible on a single field. Film usually gets damaged by the full frame. This determination is important, because the correct version of Dropout Eliminator must be used for best results. There is a version for interlaced material and a version for progressive material included in the Repair-collection. Both versions are clearly marked. The only difference between the two versions is, that the version for interlaced material holds an additional popup, where you can select the desired field to repair. The version for progressive material does not have this control and always uses the full frame.
The description below explains how to use the interlaced version. The procedure is the same for progressive material, except the field controls selection. Don't use the interlaced version for progressive video. You could, but you would have to do everything twice, because the interlaced version repairs only one field at once.
If you have more than one damage in a single frame, you can also apply the filter twice in order to repair the second damage.
Let's take a look at our damaged example:
Clearly visible is the damage in this frame. The magnification to the right shows that one field is damaged.
At first you need to apply the Dropout Eliminator plugin to the video.
To repair your video Dropout Eliminator uses a 4-step method. Just follow the four steps and the damage is gone. First you have to select the damaged frame and area. Then you have to select a frame that looks similar to the damaged one. Nearby frames usually look very similar compared to the damaged frame. So take a look around and select an appropriate frame. The next step is to modify the inserted frame to match motion and color. At last you can look at the repaired field and render it.
Step 1: Select Damaged Frame/Area
Choose "Select Damages Frame/Area" from the Step popup menu. Now you need to select the damaged frame. Move the playhead in the timeline to the damaged frame and then look in the lower left corner of in the canvas. The frame number of the selected frame is displays in the canvas. Enter the frame number into the "Damaged Frame" control.
Now you have to mark the damaged area with the four Corner controls. This can be done easily directly in the canvas. You might want to make the area a little bigger than the actual damage. If you do so, you can feather the mask in a later step to blend the replacement frame seamlessly into the damaged frame.
Step 2: Select Restoration Frame
Choose "Select Restoration Frame" from the Step popup menu. Now you need to select a nearby frame that is not damaged by choosing one of the options from the "Use Frame" popup menu.
Step 3: Modify Overlay
Choose "Modify Overlay" from the Step popup menu. Now you can modify the insertion by using the controls in this section. You don't have to do anything if your video did not change over time between the damaged and the inserted frame.
But if the camera has moved, the brightness, contrast or color did change between those two field, you can compensate the difference by doing the proper modifications in this section.
You can also find the Feather Overlay slider in this section. If the frame used for the restoration does not fit by 100%, you can feather the edges of the overlay in order to "hide" the little difference and to avoid that you will be able to see it.
Remember: Dropout Eliminator can only do a good job, when your video contains a frame that looks similar to the damaged frame. If the damage is too big and you don't have any content to insert then even Dropout Eliminator won't be able to help you. In this case it is useful to extract the damaged field and use a photo editing tool to repair the damage manually. But experience showed that usually there is enough footage that fits into the damaged area.
Step 4: Show Final
This is the last step. Look at the final result and decide if the repaired area is good to go.