Google Groups no longer supports new Usenet posts or subscriptions. Historical content remains viewable.

AI Video Bundle Review | The Best Video Creation For Beginners and Experts!

Skip to first unread message

Rabia Ken

Jan 30, 2024, 4:08:52 AMJan 30
Software programs such as Adobe Photoshop and Lightroom, DxO Optics Pro, Capture One, and others offer an extensive range of adjustments that you can make to your artworks to achieve the desired level of detail. But what if you just want high-quality results with the least amount of hassle and you have a batch of photographs to process? That's where the newest product from Topaz Labs, the AI Video Bundle, comes in.

See Here:

Instead of focusing on being as complex as other photo editing programs, Photo AI aims to handle the tedious job for you. It accomplishes this with the help of an ingenious Autopilot feature that examines your photos, recognizes the subjects and areas of poor image quality, and then uses that knowledge to execute smart adjustments with the least amount of user input. The well-received DeNoise AI, Gigapixel AI, and Sharpen AI programs from Topaz are the source of the underlying technologies.

Features of the AI Video Bundle

Compatible with pre-rendered JPEGs and Raw pictures

produces JPEG, TIFF, or PNG files that are ready for use, or DNG Raws for further processing.

able to run independently or as a plugin for programs like Photoshop

finds your subjects and automatically detects flaws in the quality of the image

rectifies lens aberrations, noise, and blurriness

employs AI algorithms to upsample low-resolution photos to higher resolutions.

finds people in your photos and enhances the clarity of the photographs

The only job left by the autopilot tool is modifying strength sliders and masks.

No subscription is required.

AI Video Bundle is priced at $199 and is now available for both Windows and Mac OS X. That's a $60 savings over the individual AI-powered Topaz apps, and it also has a more efficient workflow. Let's examine the features that AI Video Bundle provides!

a neat, straightforward UI with a few oddities

As one might anticipate from its fuss-free editing style and its rejection of features like photo management, AI Video Bundle boasts an incredibly clear and straightforward user interface. It can work as a plugin for other programs like Adobe Photoshop or as a stand-alone application. When the program is started solo, you have two options for loading images: either drag and drop them into it, or browse your disks and load them one at a time or in groups.

See Here:

When an image is imported, shot AI will automatically detect topics, faces, and poor image quality. It will then automatically produce the necessary adjustments for your shot.

Underneath a preview of the active image in the standalone app is a filmstrip of all the open photos. Strangely enough, though, the filmstrip only displays JPEG image thumbnails. Even after you've opened and examined Raw files in full size, thumbnails aren't created for them; this is something I'd like to see fixed in a later release.

Approximately 1,200 cameras, camera backs, drones, phones, tablets, and other supposedly suitable modern devices are covered by the extensive raw support. (The most notable omission is for the 'High Efficiency' Raws for the Nikon Z9, for which support is intended but not yet accessible.)

To adjust the automated corrections, a panel with an image navigator and a few other controls may be found at screen right. The mask, brush, and zoom options are located in a menu bar at the top. You can also adjust the program's preferences, such as whether it uses your CPU or any available GPUs.

Approximately 1,200 cameras, camera backs, drones, phones, tablets, and other modern devices are covered by the extensive raw support.

When you load several photos, there are two things to be aware of. To start, you have to switch between each image in the queue to start processing; the application won't process the entire queue automatically. This implies that you will have to wait a few seconds while navigating between photos during the initial load; however, after seeing each image in the filmstrip in full screen mode for the first time, all image corrections are at least remembered.

Secondly, Photo AI does not save its selected corrections to disk in a sidecar file or database; instead, it will forget them when the file or software is closed. If the image is reloaded in the future, the processing will be redone. Therefore, exporting photos with adjustments you like is worthwhile.

The resulting photos can be exported as JPEGs, TIFFs, PNGs, or DNGs, so you can choose to bake the image immediately after processing it or send a partially processed Raw file to another program, such as Photoshop, for further treatment. If your input file was raw, you can change Photo AI's default to produce a DNG Raw; if not, it will keep the original filetype.

Strong automatic subject detection, although mask refinement can be time-consuming

Hovering your mouse pointer over the subject and face boxes in Autopilot will allow you to verify the topics and faces that AI Video Bundle has detected in your snapshot. A red mask highlights the subject on the image navigator as well as the image itself, and a yellow box indicates each face that has been identified.

See Here:

I was surprised by how strong the face detection was; it could identify faces that were partially veiled or in side profile, even though they only took up a little portion of the image. You cannot deliberately choose an undiscovered face, but you can deselect the faces the algorithms found if you don't think they are significant.

When the subject of the picture is obvious, as in portraiture, the subject identification performs rather well; however, in situations where the subject blends into the background, such as in street scenes or landscapes, the selection process can be a little more haphazard. In these situations, you have the option of manually modifying the mask or changing the algorithms from "portrait" to "landscape" mode.

Instead of directly modifying your mask with a brush, you can add or subtract 'chunks' (available in four different sizes) that the algorithm has identified inside the image. This works well for small tweaks, but it can get a little laborious if you need to make big changes to the subject area. I also found myself wanting that there was a fully manual refining brush accessible.

Although autopilot reduces strain, it is missing several crucial aspects.

The Autopilot algorithms take over when subject and face detection is finished, examining the image for imperfections like noise, soft focus, motion blur, and low resolution before recommending adjustments that make sense for the particular image. Lens distortion correction is also automatically applied by default, though you can turn it off if you'd like. All things considered, Autopilot is quite simple to use and performs admirably.

See Here:

However, there is no method to adjust for exposure or white balance problems. In the aforementioned example, I was able to rectify the underexposed, cool-tinted photograph with just a few clicks using both Photoshop and PhotoLab. On the other hand, even for photos with excessive dynamic range, blown color casts, or considerable under- or overexposure, Photo AI won't automatically adjust exposure or white balance. Additionally, the user cannot directly fix these issues.

To be honest, this is the program's worst flaw, and I really hope that it will be fixed in later iterations. That essentially means that, in order to use AI Video Bundle, you still need Photoshop or a comparable application for the time being. Topaz AI, however, performs admirably with nearly no effort if your shot is properly exposed and has acceptable white balance.
0 new messages