Pocket 3 Mimo album and exports
Part of the Osmo Pocket 3 reference. Read the overview for firmware, survey scope and evidence levels.
Mimo album
Section titled “Mimo album”The camera album player displays a Low-Res label and exposes Info, Favorite, Play/Pause, Delete, Download and a scrubber. These are observed controls; opening a low-resolution preview is not downloading an original. Neither a Download tap nor an Info dialog alone proves which experiment produced the selected item. Match the file to the capture sequence before attributing its codec, dimensions or duration to a shooting-mode test.
The inspected album exposes:
| Surface | Observed controls |
|---|---|
| Library source | Device and Local |
| Filters | All, Photos, Videos, Favorites |
| Additional Local filter | Live Photo; a phone-library category, not a Pocket 3 shooting mode |
| Item grid | Media-type badges, clip duration, download icon and completed-download checkmark |
| Selection mode | Selected count/size, Cancel, Batch Select, item checkboxes, Favorites, Download and Delete |
| Batch Download Settings | Add Photo Frame toggle, Download Now, and frame styles Simple, Yearly I, Yearly II, Classic and Flagship |
For the Glamour-enabled clip, Download opened Select Format to Download with Original File, Video with Glamour Effects and Cancel. The sheet describes the first choice as retaining the original format and the second as processing videos to add effects. Both displayed an estimate of less than one minute in this case; that is a UI estimate, not measured processing time. The original branch produced the preserved file whose full hash matches the camera original. A later fresh portrait take with camera Glamour enabled completed the Video with Glamour Effects branch: Mimo showed Adding effects, then the completed-download badge. This establishes the effects download workflow separately from the local editor below. The scene contained a cat sticker and no human face, so it does not establish facial-effect efficacy or where processing runs.
Both the fresh camera original and its effects-download derivative fully decode:
| File | Complete bytes | Video encoding |
|---|---|---|
| Camera original | 50,164,533 | HEVC Main 10, 10-bit YUV420 |
| Mimo effects-download MOV | 13,871,016 | HEVC Main, 8-bit YUV420 |
Both contain 1728×3072 video at 25fps, 155 frames / 6.200s and AAC-LC 48 kHz stereo. Audio durations are 6.186667s in the original and 6.200s in the derivative. The original’s 48 contiguous HTTP 206 ranges and complete hash validated; the preserved derivative’s hash also validated. All 155 corresponding video frames differ when decoded to a common 8-bit YUV420 format, but transcoding and bit-depth conversion can contribute to those differences. This verifies a processed derivative with reduced bit depth, not a measured facial effect or camera color mode. It is not an unchanged camera-original download.
A time-aligned Local Hyperlapse item reported H.265, 3840×2160, 30FPS and 6s in Mimo Info. Those are UI properties of a local Mimo item, not an SD-file identity check. Returning from Local playback to the live Hyperlapse monitor was observed.
One settled player changed to Downloaded and removed its Download action. That is UI confirmation of Mimo’s reported transfer state; it does not by itself establish original-file integrity. Album item counts do not establish how many RAW companions or panorama component files exist. Favorite controls were exercised; after unfavorite, two settled Favorites views showed no favorite content. Persistence across a reconnect or app restart remains unverified.
Add Photo Frame was off and its style choices appeared disabled. The panel’s help distinguished standard and live photos; neither a frame export nor each style was validated. During batch download Mimo displayed transferred-item count, percentage, transfer rate and a cancel action, with pending, active and completed states on individual items. An unstable-speed/Wi-Fi-interference warning appeared during the transfer. That message is an observed app condition, not an independent diagnosis of radio interference. The first 12-item batch subsequently lost its progress banner and showed completed-download checkmarks on all selected items. This is UI completion; original metadata and companion files are separate checks. The metadata and full-file comparisons validate the 20 preserved imports. The later card inventory establishes component counts and RAW/audio companions independently of these album checkmarks. A second five-item batch also reached this UI completion state.
Mimo local editor and exports
Section titled “Mimo local editor and exports”A separate pass used Mimo’s local editor on preserved survey imports. Six MOV derivatives totaling 137,719,683 bytes were exported to the phone and preserved; all primary video and audio streams fully decode without error. Their sources match known camera originals by SHA-256. The derivatives are a separate collection from camera originals and the complete card copy.
The inspected editor exposes these UI choices:
| Control | Observed choices |
|---|---|
| Aspect | Default, 16:9, 4:3, 1:1, 3:4, 9:16, 21:9 |
| Export resolution | 720p, 1080p, 2.7K, 4K |
| Export frame rate | 30, 60 fps |
| Bitrate | Lower, Recommended, Higher |
| Noise Reduction | On/Off; Faster–Better control |
| 10-bit | On/Off |
| Color Recovery → OsmoPocket Series | D-Cinelike, D-LOG M; None also selectable |
| Portrait → Glamour Effects | Off/On; Slim, Chin, Smooth, Brighten, Enlarge, Lighten |
These are editor controls, not additional Pocket 3 shooting formats. In particular, the family’s D-Cinelike preset does not establish Pocket 3 D-Cinelike capture, and a 9:16 aspect choice or the Portrait tool does not verify native portrait recording. The local six-control Glamour panel is distinct from the eleven controls in the live-camera panel.
Three controlled export pairs all used Default aspect, 4K/30 and Recommended bitrate:
| Controlled difference | Independently inspected result |
|---|---|
| Square project, 10-bit On versus Off; Noise Reduction On/Faster | Both 2160×2160, 30fps, 241 frames / 8.033333s. On produces HEVC Main 10 / 10-bit YUV420; Off produces HEVC Main / 8-bit YUV420. |
| D-Log M source, Color Recovery D-LOG M versus None; Noise Reduction Off, 10-bit On | Both 3840×2160 HEVC Main 10, 30fps, 218 frames / 7.266667s. All 218 corresponding decoded video frames differ. |
| Square source, local Glamour master On versus Off; Noise Reduction Off, 10-bit On | Both 2160×2160 HEVC Main 10, 30fps, 153 frames / 5.100000s. All 153 corresponding decoded video frames differ; first-frame review shows changes in the face region. |
All six contain AAC-LC, 48 kHz stereo; within each pair the decoded audio is identical. The D-Log M source was 4K/25 with mono audio, so its 30fps/stereo exports also demonstrate that editor output timing and channel count can differ from the source. No new stereo spatial information or particular frame interpolation algorithm is established. Likewise, the square project’s 4K export label produced 2160×2160, not the source’s 3072×3072 dimensions.
The local Glamour pair retained the installation’s stored strengths: Slim 41, Chin 46, Smooth 78, Brighten 43, Enlarge 35 and Lighten 57. They are neither factory defaults nor tested endpoints; no strength slider was changed. Only the master state changed between exports. Pixel differences do not isolate an individual control, measure quality, identify a LUT/log curve or establish the processing location. These local-editor results are separate from the Device Download → Video with Glamour Effects workflow documented above.