AI Model Pose Library Statistics 2026: 100 Poses, 500 Angles
The Berrys pose library publishes 100 commercial poses as 500 camera-angle records, and 500 of those records (100%) ship a ControlNet depth map. 426 carry a written pose brief (85.2%); the remaining 74 are visual-only. This page counts that dataset directly, so every number below can be reproduced from the public JSON.
Counted from lib/poses_full_sfw.json · Last updated 6 November 2026 · Dataset at /poses-full.json · Markdown version
The library in five numbers
Key findings
Each finding is self-contained and anchored, so it can be quoted or linked on its own. Share denominators are stated with every figure.
All 500 of 500 camera angles (100%) ship a paired ControlNet depth map, a line-art mannequin render and a JPEG thumbnail, so every pose in the library can drive pose-conditioned generation without a photoshoot.
500 depth maps / 500 angle recordsLink to this findingThe library holds 100 distinct poses across 7 posture families. Hand Gestures is the largest at 27 poses (27% of poses, 135 of 500 angles), followed by Seated at 23 poses (23%).
100 poses categorisedLink to this findingEvery pose resolves to exactly 5 camera angles — Front, Top-Down, Low-Angle, Side, 3/4 View — and all 100 of 100 pose records (100%) carry the complete set, which is what lets one garment be rendered as a 5-view rotation instead of a single hero frame.
100 complete pose records / 100 posesLink to this finding426 of 500 angle records (85.2%) carry a written pose brief averaging 17.5 words, while 74 (14.8%) are visual-only and must be described from the depth map before prompting.
426 briefed / 500 angle recordsLink to this findingThe 426 written briefs collapse to 98 distinct descriptions, because the same posture text is reused across that pose's five camera angles — one description per pose, not one per view.
98 distinct texts / 100 posesLink to this finding73 of 100 poses (73%) belong to the Commercial Catalogue Series series with full-body framing, and 27 (27%) to Upper Body Framing for crops that keep the legs out of frame.
100 poses by framing seriesLink to this findingBreakdown by posture family
| Posture family | Poses | Angles | Share of poses | Best for |
|---|---|---|---|---|
| Hand Gestures | 27 | 135 | 27% | Torso-up crops, hold-the-product frames, accessories and small electronics |
| Seated | 23 | 115 | 23% | Bottoms, footwear on a low stool, home and lifestyle SKUs |
| Standing | 17 | 85 | 17% | Full-length apparel, outerwear, bags, full-body catalogue covers |
| Kneeling | 12 | 60 | 12% | Denim, activewear, mid-length dresses, ground-plane product placement |
| Floor Recline | 9 | 45 | 9% | Swimwear, loungewear, bedding and home textiles |
| Crouch | 6 | 30 | 6% | Sneakers, sports accessories, streetwear lookbooks |
| Forward Lean | 6 | 30 | 6% | Torso-up beauty and skincare, shelf and counter contexts |
What ships with every camera angle
| Asset | Records | Coverage | Path pattern |
|---|---|---|---|
| ControlNet depth map (PNG) | 500 | 100% | /for/poses/img/depth/full_single_P001_front.png |
| Line-art mannequin render (PNG) | 500 | 100% | /for/poses/img/line/full_single_P001_front.png |
| Thumbnail (JPEG) | 500 | 100% | /for/poses/img/thumb/full_single_P001_front.jpg |
| Written pose brief | 426 | 85.2% | brief field in /poses-full.json |
The five-view angle set
100 of 100 pose records (100%) carry all 5 views. That completeness is what makes a rotation possible: one pose record can produce a comparable multi-view set instead of a single hero frame.
Flat-on framing for cover images and primary gallery slots.
Top-down framing for flat-lay contexts and table-top SKUs.
Low angle for hero elevation and taller silhouettes.
Profile view for drape, hem line and bag depth.
Three-quarter view, the marketplace default for perceived depth.
How to use this data in a canvas workflow
Choose a posture family from the table by product type, then note the pose code (P001–P100). The code is the stable key across the gallery, the JSON and the image paths.
Fetch the depth PNG for the view you need and drop it on the canvas as the pose reference next to your File Node packshot. Depth pins limb position, which is what stops the model inventing a new stance.
Read the written brief when it exists. Where it is empty (74 of 500 records), describe stance, weight-bearing leg and hand position from the depth map before prompting.
With the Brand System Node feeding the image node, lighting, palette and lens character stay identical across all 5 views — the difference between a lookbook and five unrelated renders.
Method
| Source | lib/poses_full_sfw.json, published verbatim at https://berrys.io/poses-full.json (100 pose records). |
|---|---|
| Counting unit | Pose records (100) and camera-angle records (500). One pose record always contains 5 angle records. |
| Coverage test | An asset counts as present only when its path field is a non-empty string. Empty or missing values count as absent, never as inferred. |
| Deduplication | Identical brief text is collapsed before counting distinct descriptions, which is why 426 briefed records reduce to 98 unique texts. |
| Exclusions | No generated images, no model output, no marketplace compliance data. This study counts the reference dataset only. |
| Rounding | Shares are rounded to one decimal place. Gaps under one percentage point are not called differences. |
| Refresh | Recomputed whenever the dataset is regenerated. Last updated 6 November 2026. |
Cite this page
Use this citation when quoting a figure, and link the anchor if you quote a single finding.
FAQ
How many poses are in the Berrys pose library?
100 poses, published as 500 camera-angle records — 5 angles per pose — with 500 ControlNet depth maps.
Link to this answerDo all poses include a ControlNet depth map?
Yes. 500 of 500 angle records (100%) ship a depth map, plus a line-art render and a thumbnail, so any pose can be used as a pose-conditioning reference.
Link to this answerHow many poses have a written prompt brief?
426 of 500 angle records (85.2%) carry a written brief averaging 17.5 words. The other 74 are visual-only, so describe the geometry from the depth map instead of the thumbnail.
Link to this answerWhich pose family has the most options?
Hand Gestures, with 27 poses and 135 of 500 angle records (27%).
Link to this answerIs the pose library free to use commercially?
Yes. The library is free to browse, copy and reference, and the JSON catalog is public at https://berrys.io/poses-full.json. Platform rules still apply to the finished image — a marketplace-compliant render depends on background, aspect ratio and frame fill, not on the pose reference.
Link to this answerRun the workflow, not the reshoot.
The pose, the depth map and the brand rules live on one canvas. Load a reference, keep the Brand System Node upstream, and generate the full 5-view set without a studio booking.