Picturesque 0.1.1

A friendlier first run, and fixes for model downloads.

Picturesque only ever copies your photos. It reads your original folders and copies photos into its own library (output) folder; originals are never moved, renamed, edited or deleted.

New

• Setup wizard on first launch: choose the output (library) folder and its layout, add the folders to import from, and optionally download the AI models. Reopen it any time from Settings → Run setup again. (#144)
• Culling tour: the first time your library has photos, a short tour points at the timeline, the pick/reject/rate keys, the loupe, Compare, Filters and Export. Replay it from Settings. (#147)
• Help and keyboard shortcuts: press ? (or click Help) for every shortcut and links to the guides. (#146)
• Clearer empty screens: each one says what to do next and takes you there. (#145)
• "AI analysis needs models": the library says which analyses wait for a model download instead of silently doing nothing. (#140)

Fixed

• Large AI models failed to download on slow connections after 60 seconds; each file may now take up to an hour. (#137)
• Links (guides, model licenses) now open in your browser from the desktop app. (#146)

Install

System	File
Windows 10+ (64-bit)	Picturesque_0.1.1_x64-setup.exe (no admin rights), or the .msi for all users
macOS (Apple Silicon)	Picturesque_0.1.1_aarch64.dmg
Ubuntu / Debian	sudo apt install ./Picturesque_0.1.1_amd64.deb
Fedora	sudo dnf install ./Picturesque-0.1.1-1.x86_64.rpm
Other Linux	Picturesque_0.1.1_amd64.AppImage (chmod +x, then run)

The installers are not signed yet. Windows: if SmartScreen warns, choose More info → Run anyway. macOS: if the app won't open the first time, go to System Settings → Privacy & Security and click Open Anyway (macOS support is still experimental, #52). On Windows, HEIC photos can't be shown or analyzed yet (#135).

Check downloads against SHA256SUMS.txt. Full guide: Installing.

What's Changed

• #137 Allow an hour to download each model file by @denyszorinets in #138
• #139 Release workflow: flatten installers before checksums by @denyszorinets in #141
• #140 Say that AI analysis needs models when they are not downloaded by @denyszorinets in #142
• #144 First-launch setup wizard by @denyszorinets in #148
• #146 Keyboard shortcuts overlay and help links by @denyszorinets in #149
• #145 Empty states that go straight to the next step by @denyszorinets in #150
• #147 Guided culling tour by @denyszorinets in #151
• Release 0.1.1 by @denyszorinets in #152

Full Changelog: v0.1.0...v0.1.1

Picturesque 0.1.0

First release of Picturesque. Download the installer for your system below.

System	File	Install
Windows 10+ (64-bit)	Picturesque_0.1.0_x64-setup.exe	Run it (no admin rights needed). Or Picturesque_0.1.0_x64_en-US.msi to install for all users.
macOS (Apple Silicon)	Picturesque_0.1.0_aarch64.dmg	Open the dmg and drag Picturesque to Applications.
Ubuntu / Debian	Picturesque_0.1.0_amd64.deb	sudo apt install ./Picturesque_0.1.0_amd64.deb
Fedora / rpm-based	Picturesque-0.1.0-1.x86_64.rpm	sudo dnf install ./Picturesque-0.1.0-1.x86_64.rpm
Other Linux	Picturesque_0.1.0_amd64.AppImage	chmod +x Picturesque_0.1.0_amd64.AppImage and run it.

The installers are not signed yet:

• Windows: if SmartScreen warns you, choose More info → Run anyway.
• macOS: if macOS won't open the app the first time, go to System Settings → Privacy & Security and click Open Anyway next to Picturesque. macOS support is still experimental (#52).

Known issue: on Windows, HEIC photos (from iPhones) are imported but can't be shown or analyzed yet (#135).

You can check your download against SHA256SUMS.txt. Full guide: Installing.

What's Changed

• #1 Update devcontainer for Rust + Tauri with libvips/LibRaw by @denyszorinets in #2
• #4 Set up Sphinx documentation site by @denyszorinets in #56
• #5 Write architecture overview and core ADRs by @denyszorinets in #57
• #6 Scaffold Cargo workspace by @denyszorinets in #58
• #7 Scaffold Tauri 2 + Svelte 5 app by @denyszorinets in #59
• #8 Add dev HTTP server and browser transport by @denyszorinets in #60
• #9 Set up CI on GitHub Actions by @denyszorinets in #61
• #10 Spike: image decoding stack by @denyszorinets in #62
• #63 Add HEVC decoder to devcontainer by @denyszorinets in #64
• #12 Define photo domain model by @denyszorinets in #65
• #13 Create SQLite catalog with migrations by @denyszorinets in #66
• #14 Add library settings by @denyszorinets in #67
• #15 Implement source folder scanner by @denyszorinets in #68
• #16 Implement metadata reader by @denyszorinets in #69
• #17 Implement offline reverse geocoder by @denyszorinets in #70
• #18 Implement path template engine by @denyszorinets in #71
• #19 Implement import use case by @denyszorinets in #72
• #20 Build import UI by @denyszorinets in #73
• #22 Build thumbnail and preview cache by @denyszorinets in #74
• #23 Serve images to the UI by @denyszorinets in #75
• #24 Add catalog query API by @denyszorinets in #76
• #25 Build timeline grid by @denyszorinets in #77
• #26 Build loupe view by @denyszorinets in #78
• #27 Implement keyboard culling by @denyszorinets in #79
• #28 Build filter panel by @denyszorinets in #80
• #29 Write XMP sidecars by @denyszorinets in #81
• #31 Build background analysis queue by @denyszorinets in #82
• #32 Implement technical quality scores by @denyszorinets in #83
• #33 Add perceptual hash similarity by @denyszorinets in #84
• #34 Implement series detection by @denyszorinets in #85
• #35 Rank best photo in each series by @denyszorinets in #86
• #36 Show series frames in the grid by @denyszorinets in #87
• #37 Build compare view by @denyszorinets in #88
• #38 Add score filters and sorting by @denyszorinets in #89
• #40 Integrate ort with execution providers by @denyszorinets in #90
• #41 Build model manager by @denyszorinets in #91
• #43 Compute CLIP image embeddings by @denyszorinets in #92
• #42 Add inference benchmark (part 1) by @denyszorinets in #93
• #44 Add aesthetic score by @denyszorinets in #94
• #45 Add semantic text search by @denyszorinets in #95
• #46 Detect faces and closed eyes by @denyszorinets in #96
• #47 Combine scores for ranking and series by @denyszorinets in #97
• #48 Show AI results in the UI by @denyszorinets in #98
• #50 Move rejects to a folder by @denyszorinets in #99
• #51 Export picks by @denyszorinets in #100
• #55 Write user documentation by @denyszorinets in #101
• #52 Build Linux installers by @denyszorinets in #102
• #54 Smoke-test the real app by @denyszorinets in #103
• #53 Add release workflow by @denyszorinets in #104
• #105 Persistent import pipeline by @denyszorinets in #110
• #106 Watched import folders by @denyszorinets in #111
• #107 Activity view; remove import runs by @denyszorinets in #112
• #108 Smoke-test continuous import in the real app by @denyszorinets in #113
• #52 Make targets for installers by @denyszorinets in #114
• #115 Pause importing and analysis when a disk is full by @denyszorinets in #116
• #117 Cull from the loupe; mouse multi-select by @denyszorinets in #118
• #119 Auto-pick keepers by @denyszorinets in #120
• #121 Show the right photos right after filtering by @denyszorinets in #122
• #123 Statistics: keepers per camera and lens, focal lengths per camera by @denyszorinets in #124
• #125 Camera filter as one-click buttons near the top by @denyszorinets in #127
• #126 Re-analyze already imported photos by @denyszorinets in #128
• #129 One row and one filter option per camera name by @denyszorinets in #130
• #131 Windows installers; green Windows and macOS CI by @denyszorinets in #132
• #133 Website: landing page and docs on GitHub Pages by @denyszorinets in #134
• Release 0.1.0 by @denyszorinets in #136

New Contributors

• @denyszorinets made their first contribution in #2

Full Changelog: https://github.com/denyszorinets/picturesque/commits/v0.1.0