Сортировочная ячейка Isaac Sim: CV-пайплайн и меши товаров

Замкнутый контур "поток -> CV -> механика": товары идут по конвейеру с шагом 700 мм,
класс определяется стереопайплайном во время движения, пушер и плуг реагируют физически.

Состав:
* control_test/ - ячейка и CV. run_sorting_cv.py + cv_worker.py (два процесса, потому что
  torch внутри Isaac роняет сцену), cell.py (физика лент, плуга, пушера), measure_plane.py
  (замер габаритов), README.md и .memory.md с замерами, проблемами и ловушками
* robozon_sorter/ - модули симуляции, scripts/ - утилиты, scene/ - сцены
* assets/ - меши товаров, плуг, объекты Objaverse

Бейзлайн CV: DEFOM-Stereo vitl, вход 480, iters 24, кроп зоны осмотра, без сегментации.
На потоке 700 мм - классы 8/9, габариты MAE 32.8 мм, 469 мс на товар при такте 700 мс.

Веса моделей (4.5 ГБ) и пропсы конвейера NVIDIA (274 МБ) не включены - источники и
команды скачивания в MODELS.md. Выход прогонов (captures/, runtime/) не включён:
воспроизводится.

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>

.gitignore

@@ -0,0 +1,25 @@
+# Веса моделей в репозиторий не кладутся - ссылки на источники в MODELS.md
+*.pth
+*.pt
+*.onnx
+*.ckpt
+*.safetensors
+
+# Пропсы конвейера NVIDIA (274 МБ): не наши и скачиваются штатным скриптом
+assets/conveyors/
+# восстановить: python3 scripts/fetch_assets.py
+
+# Выход прогонов - воспроизводится, в истории не нужен
+control_test/captures/
+control_test/runtime/
+control_test/diag/
+runs/
+
+# Резервные копии, которые делались перед правками
+*.bak
+*.orig
+*.pre*
+
+__pycache__/
+*.pyc
+.DS_Store

CLAUDE.md

@@ -0,0 +1,328 @@
+# Context for Claude
+
+Working notes for this repo: how the cell is put together, where the rest of the assets
+live, and the traps that are not visible from the code. Read this before changing anything
+in `robozon_sorter/sim/` — several constants here are load-bearing and look arbitrary.
+
+---
+
+## 1. What this is
+
+A conveyor sorting cell in Isaac Sim 6.0 (Kit 110). Goods ride a belt, three stereo pairs
+measure each one under a camera portal, and a pusher diverts the round ones onto a branch
+belt and into a bin.
+
+```
+infeed belt ──▶ camera portal ──▶ laser gate ──▶ pusher ──▶ Belt_01 ──▶ bin
+   x=+2.3          x=-0.75          x=-3.74      x=-3.90              y=+1.3…2.0
+                (CRE-ROI v2b)                  (class D only)
+```
+
+Goods travel in **−X** at 1 m/s. Belt surface is at **z = 1.781** everywhere on the main
+run and the branch.
+
+**Classes:** B = sortable, C = oversize (any dim >450×320×320 mm or <10 mm), D = round
+(`K = r_in/R_out > 0.8`). Only D is diverted.
+
+---
+
+## 2. Where things are
+
+### This repo
+`scene/sorter.usd` is the real authored cell, not a procedural rebuild. It was exported
+from `90_degree.usd` with two changes: conveyor references re-pointed to
+`../assets/conveyors/`, and the 48-object library dropped in favour of six bundled test
+meshes with baked geometry.
+
+Do not "simplify" the scene by regenerating it from code. That was tried and rejected —
+the result looked nothing like the real cell.
+
+`scene/plow_cell.usd` is the same authored build carried over whole, but as the **bare
+mechanics**: conveyors, the Y-split pusher and the plow, with no camera portal, no laser
+gate and no item library. Cameras, speed scenarios and laser sensors get added on top of
+it. Rebuild it with `scripts/build_plow_cell.py`, which re-points the S3 conveyor
+references at `assets/conveyors/`, points the plow at `assets/plow/`, and strips the baked
+`targetPosition.timeSamples` track off the pusher drive — time samples outrank the
+attribute default, so while they are present nothing can *control* that drive.
+
+The plow meshes matter: `assets/meshes/plow_*.usd` are 8-point placeholder boxes that
+`sorter.usd` still uses, while `assets/plow/plow_*.usd` is the real geometry (72k / 23k
+points). `scripts/smoke_plow_cell.py` asserts the point count precisely because a stub
+composes without error and just looks wrong.
+
+### Remote machine (where Isaac Sim runs)
+
+```
+host      dasha@46.39.224.77
+isaac     /home/whatevenif/isaacsim          (python.sh lives here)
+project   /home/dasha/robozon-sorter          (deployed copy, ~308 MB, runnable)
+assets    /home/dasha/isaac_assets            (the original working tree)
+```
+
+Connect with an SSH ControlMaster (plain repeated ssh exhausts local ephemeral ports on
+long sessions) and tunnel the Kit python server:
+
+```bash
+ssh -M -S ~/.ssh/cm/dasha -o ControlPersist=900 -fN dasha@46.39.224.77
+ssh -fN -L 8226:127.0.0.1:8226 dasha@46.39.224.77       # isaacsim.code_editor.python_server
+```
+
+Then send code with the `isaac-sim-remote` skill's `isaacsim_send.py`. Named contexts keep
+state between calls, which is how the long experiments were run.
+
+### More meshes, if six are not enough
+
+| What | Where | Notes |
+|---|---|---|
+| 31 classified objects | `/home/dasha/isaac_assets/categories.json` | 13 C, 11 D, 7 B; has `zone`, `obb_extents_m`, `k_round`, `label_ru` |
+| the objects themselves | `/home/dasha/isaac_assets/robozon_conveyor_scaled.usd` under `/World/CVObjects` | 48 prims, 31 of them classified |
+| raw source geometry | `/home/dasha/.objaverse/hf-objaverse-v1/glbs/` | 19 `.glb`, 123 MB — what CVObjects reference |
+| already-exported | `/home/dasha/isaac_assets/export_meshes/` | flattened USDs, the six in this repo came from here |
+| conveyor art | `/home/dasha/isaac_assets/Props/Conveyors/` | 283 MB, mirrors the Omniverse S3 tree |
+
+To add a test mesh, flatten it out of the source scene so it carries no `.glb` reference:
+
+```python
+ns = Usd.Stage.CreateInMemory()
+root = UsdGeom.Xform.Define(ns, "/Item")
+root.GetPrim().GetReferences().AddReference(
+    "/home/dasha/isaac_assets/robozon_conveyor_scaled.usd", f"/World/CVObjects/{name}")
+ns.SetDefaultPrim(root.GetPrim())
+ns.Flatten().Export(f"assets/meshes/{name}.usd")
+```
+
+Then add an entry to `assets/meshes/manifest.json` with `zone` and `gt_dims_mm`. Objects in
+that scene sit at **0.49× real size** (1/3-scale meshes × 1.4706), which is why
+`config.DIM_SCALE = 1/2.041` converts metres to real millimetres.
+
+### Scene backups
+`/home/dasha/isaac_assets/backups/` — `robozon_prescale_2252.usd` is the clean pre-scale
+original, useful if the working scene ever gets damaged.
+
+---
+
+## 3. Mechanics
+
+`sim/scene.py` opens the USD and re-applies the runtime settings USD does not carry.
+`sim/mechanics.py` is the per-item behaviour. `sim/spawner.py` hooks a PhysX step callback
+so the cell runs itself from the Play button.
+
+### Belts
+Driven by `PhysxSurfaceVelocityAPI` on kinematic slabs, not by the authored
+`ConveyorBeltGraph` nodes — those carry no velocity and only fight the explicit setting, so
+they are switched off on load.
+
+**`surfaceVelocity` is expressed in the body's LOCAL frame.** `Belt_01`
+(`/World/ConveyorTrack_03/Belt_01`) is rotated: its local X points along world −Y. Carrying
+goods toward the bin therefore needs `(−speed, 0, 0)`. Setting the intuitive `(0, +speed, 0)`
+drags them sideways and they sit there looking stuck. Resolve the local axes with
+`XformCache.GetLocalToWorldTransform(prim).TransformDir(...)` before setting it.
+
+### Pusher
+A kinematic blade moved directly from script. It is **not** joint-driven: the authored
+`PusherSlide` prismatic drive is unusable because USD drive-target writes reach PhysX about
+a second late, so the blade never finishes its stroke while the item is still in reach.
+The joint is disabled and `DiverterAnimGraph` is switched off (it rewrites diverter targets
+every tick).
+
+Blade collisions are filtered against every belt so it sweeps through the side rails
+instead of grinding on them.
+
+**Speed and physics rate are a matched pair: 2.5 m/s at 120 Hz.** Do not change one alone.
+
+* The blade covers 0.70 m of belt; the beam trips with the item at x = −3.61, so the whole
+  cycle must fit in 0.64 m of travel = 0.64 s at 1 m/s.
+* Stroke is 0.72 m → extension alone needs >1.12 m/s, extend+retract >2.25 m/s.
+* Measured: 1.0 / 1.5 / 2.0 / 2.5 m/s all deliver to the bin. **3.0 m/s throws the item**
+  (ends up at y≈122, z≈−2070) — the kinematic blade injects too much impulse.
+* Raising the solver rate to 480 Hz to "smooth" the impulse also breaks the landing. It was
+  tried; the item stopped reaching the bin. 120 Hz is the validated setting.
+
+
+### Plow sorting station (`plow_cell.usd`)
+
+Two-way sort at the plow, verified by measurement. Rebuild the geometry with
+`scripts/narrow_plow.py` then `scripts/place_plow_lanes.py`.
+
+**Arm width is 600 mm**, not the authored 730. The wider arm overhung both rails and
+clipped goods it should have passed. `narrow_plow.py` does not guess which local axis
+carries the length - it tries each and measures; the answer is **local Y, scale 0.8219**.
+Hinge, drive and limits untouched.
+
+**Lane geometry, all verified against the belt (main belt x -7.00..-6.00, y +-0.45, top
+z 1.7805):**
+
+| | span | contact |
+|---|---|---|
+| lane B, perpendicular, travel -Y | x -7.03..-6.58, y -2.45..-0.45 | height 0.0 mm, Y 0.0 mm |
+| lane C, 45 deg, travel (-X,+Y) | x -8.12..-6.39, y +0.45..+2.18 | height 0.0 mm, Y 0.1 mm |
+
+Three placement mistakes were made and corrected, all of which looked fine in the tree and
+wrong in the viewport:
+
+1. **Height.** These tracks carry an authored `-0.1` z offset, putting their belts at
+   z 1.681 - 100 mm below the run, so they read as separate furniture. Use `LANE_DROP = 0`.
+2. **X position.** Lane B was first placed at x -7.48..-7.03, entirely *past* where the
+   main belt stops (-7.00). A plow sweeps goods sideways while they are still on the belt,
+   so a lane must run **alongside** it inside the arm's span (x -7.12..-6.52), never beyond
+   the end. Same for C: at x -6.90 its near corner sat behind the plow and +Y deflections
+   had nothing to land on; -6.55 puts the corner at the arm tip.
+3. **A 45 deg lane does not meet a straight edge at its centreline.** Its near corner runs
+   ahead by 159 mm (measured), so the lane needs that much offset or it cuts into the belt.
+
+**An angled lane needs a transition DECK, not a corner patch.** Offsetting by 159 mm makes
+its near corner touch, but only that one corner - the other stands off by
+width/sqrt2 = 318 mm. Patching just that triangle is not enough either: the plow can put a
+item anywhere across the discharge width, so anything pushed wide still drops through the
+second gap. `place_plow_lanes.add_transition()` therefore decks the whole corner - the
+convex span of the belt edge across the junction (x -7.00..-6.00) and both end-face
+corners, giving y 0.45..0.768 - coplanar with both belt surfaces. A square lane like B
+meets flush along its whole face, so the builder returns nothing rather than emitting a
+zero-area collider.
+
+**Every lane also needs a corner deck**, flush or not. The belt is wider than the lane -
+the run reaches x=-6.00 while lane B stops at -6.58 - so the right angle between them is
+open air, and anything the plow pushes sideways in that leftover span drops through.
+`add_corner_deck()` fills it with a triangular fillet (B: legs 575 mm, corners
+(-6.575,-0.45), (-6.00,-0.45), (-6.575,-1.025); C: legs 391 mm), turning the right angle
+into a chute. Deck plus fillet together leave no open surface across the discharge.
+
+Lanes start at y = +-0.45, which clears the arm's swept envelope: a 600 mm arm at +-35 deg
+reaches 0.60*sin(35) = **0.344 m** either side.
+
+**The conveyor shell collides.** `SM_ConveyorBelt_*_02` has `collision=True`, and that
+includes the blue side rails - they physically block goods from leaving the belt, which is
+exactly what "nothing reaches the bins" looks like. `plow_sort.open_junction()` clears the
+collider on the three shells at the junction, the way a real plow station has its rails cut
+away. Each `Belt` keeps its own collider, so nothing falls through.
+
+**Plow delivery is geometrically impossible as currently placed - measured.** The arm is
+600 mm and pivots at the belt centre (-7.05, 0), so its lateral reach is
+0.60*sin(35) = **0.344 m**. Both lanes start at **y = +-0.45**. That leaves a **106 mm dead
+band**: the blade can push an item to 0.344 and no further, the lane begins at 0.45, and the
+item runs off the belt end and stops at x ~= -7.0 with v = 0, or falls. Every trace in
+`runs/gt_run5.json` shows exactly that - barrel drifts from y=-0.09 to y=+0.148 and halts.
+
+The arm sits at z 1.810..1.890 while the lane surfaces are at 1.7805, i.e. **29 mm above
+them**, so it would sweep *over* a lane edge rather than into it. The earlier reasoning that
+lanes had to start beyond the swept envelope was therefore unnecessary, and it is what
+opened the dead band. Closing it means one of: lanes in to ~y=0.33, a longer arm, or a
+larger swing (the joint limit is +-35 authored).
+
+**The angle does not track the command.** Measured in a demo run: an item commanded 0.0 saw
+the arm at **+30.5**, one commanded +30.0 saw **-34.6**. Both then went to the same lane. So
+on top of the reach gap the control sign and the settling are wrong. Two things are known
+about this path and both matter:
+
+* `Plow.target(deg)` writes the whole angle at once; calling it *and* `step_toward()` leaves
+  the ramp nothing to do and the blade snaps. Use one or the other, and `plow_sort` uses the
+  ramp.
+* `Plow.home()` does not settle the arm - it sets a target, and the compliant drive needs
+  time. Consecutive trials therefore start from wherever the previous one left the blade
+  (measured: -26.4 and -14.4 at the start of runs meant to begin at 0).
+
+* A **positive command deflects to -Y (lane B)**, opposite to the natural reading. Measured,
+  not assumed, exactly as the module docstring warns.
+
+Until reach and tracking are fixed the plow does not deliver: `scripts/run_demo.py` reports
+the classification honestly and a delivery rate near zero.
+
+### Retract interlocks
+The blade returns only when (a) the pushed item has cleared to y > 0.5 and (b) no other item
+is inside the blade's footprint. Retracting blindly sweeps the blade back through the next
+item and knocks it over — that was a real observed failure, not a hypothetical.
+
+### Plow (`plow_cell.usd` only)
+The second diverter, at x = −7.05. Mechanically the opposite of the pusher: a **dynamic
+arm on a revolute joint with an angular force drive** (axis Z, limits ±35°, stiffness
+120000, damping 1500, 12 kg with gravity disabled). Being force-driven it is compliant —
+it yields on contact instead of teleporting through cargo — so `sim/plow.py` commands a
+drive *target* rather than writing a transform.
+
+Two consequences: a target is a request, not a position (read `Plow.angle`, which measures
+the arm's simulated pose, never assume it arrived), and the rate is not free. The authored
+graph swings 30° in **7 ms** (72 rad/s), which is a display animation, not a sortable
+motion — at that rate the blade lands as an impulse, the same failure the pusher shows
+above 2.5 m/s. `Plow.step_toward` ramps the target at `config.PLOW_RATE` (180 °/s) instead;
+`config.PLOW_RATE_AUTHORED` keeps the original figure for reference.
+
+The scene keeps its authored `DiverterAnimGraph`, so pressing Play alone demonstrates the
+cell. `plow_cell.prepare(..., script_control=True)` switches that graph off — it has to go,
+or it rewrites the drive target every tick and overwrites anything Python commands.
+
+There is also a stray second `ConveyorTrack_01` at stage root, outside `/World`, left over
+from how the cell was assembled; it composes as a duplicate belt in the same place and is
+deactivated on load rather than deleted, so the file stays as authored.
+
+### Laser gate
+A genuine `raycast_closest` across the belt, not a coordinate test. The beam starts at
+y = −0.24, which is deliberately **clear of the blade's retracted footprint** (the blade
+spans y −0.33…−0.27). Start it any further out and the ray simply reads the blade and the
+gate never sees cargo.
+
+---
+
+## 4. Vision (CRE-ROI v2b)
+
+`cv/pipeline.py`. Per item, once, while it sits under the portal: FastSAM segment-everything
+→ keep the blob covering the projected inspection point in every view → ROI crop to a fixed
+320 px side → **one batched CRE-Stereo pass over all three crops** → fuse, dropping views
+whose 3D centroid disagrees with the median by >10 cm → belt-plane OBB + `r_in/R_out`.
+
+The crop must use the **identical column window in both eyes**, left-padded by the maximum
+disparity, or the right-hand counterpart falls outside the crop. Disparity is invariant to
+an equal column shift, so depth stays correct.
+
+**The stereo rig must be rectified.** Both eyes of a pair share one orientation, with the
+right eye offset along its X. Aiming each eye separately at the target verges the pair and
+breaks `depth = fx·b/disp` — reconstruction came back at 2–5 m instead of 0.63 m. If
+`config/calib.json` ever loses its `"rectified": true` flag, the pipeline refuses to load
+rather than silently producing garbage.
+
+Measured: dimensions land within ~34 mm of ground truth on the largest edge; the batched CRE
+pass costs ~170 ms per item.
+
+---
+
+## 5. Traps that cost real time
+
+Each of these presented as a silent failure, not an error.
+
+- **Copied items arrive kinematic and hidden.** Kinematic bodies ignore gravity and belt
+  friction; hidden ones are invisible to the cameras while still simulating. Both must be
+  cleared, plus `sleepThreshold = 0` or a settled item is never woken by the belt.
+- **`BBoxCache` / `XformCache` return the AUTHORED transform during simulation.** A moving
+  item looks frozen. Use `RigidPrim.get_world_poses()`.
+- **`timeline.stop()` resets items to their authored poses**, so captures of a finished run
+  must be taken while still playing.
+- **`rep.orchestrator.step_async()` stops the timeline.** Any live loop that classifies must
+  call `timeline.play()` again afterwards or the whole line freezes mid-run.
+- **The viewport's active camera may not be Persp.** Setting Persp's pose then does nothing;
+  call `vp.set_active_camera("/OmniverseKit_Persp")` first.
+- **Camera bodies carry cosmetic `AimRay` cones** that sit right over the inspection point
+  and dominate the frame. Hidden on load.
+- **Stale composition:** if a stage was opened while a referenced asset was missing, USD
+  caches the failure. Dropping the file in later does not fix it — the prim stays valid with
+  `localErrors: none` but empty typeName and bbox. Diagnose with `GetPrimStack()`, fix by
+  re-opening the stage.
+- **`Stage.TraverseAll()` crashes Isaac on this scene.** Use targeted `GetPrimAtPath`, or run
+  read-only checks headlessly via `/home/whatevenif/isaacsim/python.sh` (has pxr, no Kit).
+- **Cube colliders:** use `size = 2.0` so the scale op equals the half-extent. Any other
+  arrangement makes PhysX use the wrong bounds and goods drop through the belt.
+
+---
+
+## 6. Known limitations
+
+- **Classification is weak.** Metrology is sound but the roundness metric does not separate
+  classes at this scale: genuinely round items read K ≈ 0.75 against a 0.80 threshold while a
+  plain box reads 0.76. `config.ROUND_K` needs recalibration before the vision output should
+  drive the pusher for real. Use `--no-vision` to exercise mechanics on ground truth.
+- **Consecutive class-D items are missed at the default 700 mm pitch.** Headway is 0.70 s and
+  one pusher cycle is 0.62 s; the gate is blind while a cycle runs, so a D item immediately
+  behind another D crosses the beam unseen. Verified: `D,C,D,B` sorts both D correctly, `D,D`
+  back-to-back loses the second. Fixing it needs a larger pitch, a slower belt or a second
+  diverter — not a faster blade, which is already at its stability limit.
+- **The plow (`DiverterEnd`) is untouched in `sorter.usd`.** There its geometry was
+  restored and its arm returned to rest, but its control logic is left exactly as authored.
+  It is `scene/plow_cell.usd` that makes the plow controllable — see below.

MODELS.md

@@ -0,0 +1,65 @@
+# Веса моделей
+
+В репозиторий не попадают: около 4.5 ГБ, и все скачиваются из первоисточников. Ниже —
+откуда именно, чтобы окружение поднималось повторяемо.
+
+## DEFOM-Stereo — рабочий бейзлайн пайплайна
+
+Репозиторий: <https://github.com/Insta360-Research-Team/DEFOM-Stereo>
+Веса: <https://drive.google.com/drive/folders/1cZLcIjHlmUo986gkR6FbofG1cj5BT36x>
+
+```bash
+git clone https://github.com/Insta360-Research-Team/DEFOM-Stereo.git
+cd DEFOM-Stereo && mkdir -p checkpoints && cd checkpoints
+gdown 1XuAM4vqzura_6NKN70hMW5lFD4TafnDL   # defomstereo_vitl_sceneflow.pth  1.53 ГБ  <- используется
+gdown 1qyXKO-Nxq3ndl2H0deQpo6BSvwlGKYEg   # defomstereo_vits_sceneflow.pth  173 МБ   (лёгкий вариант)
+wget https://huggingface.co/depth-anything/Depth-Anything-V2-Large/resolve/main/depth_anything_v2_vitl.pth
+```
+
+Энкодер DepthAnythingV2 обязателен: без него модель не поднимается.
+
+Скачивание папкой целиком (`gdown --folder`) не работает — перечисляет файлы и ничего не
+качает. Только по одному идентификатору.
+
+В репозитории есть и штатные скрипты: `scripts/download_models.sh`, `scripts/download_dav2.sh`.
+
+## CREStereo — второй движок, точнее по габаритам
+
+<https://github.com/megvii-research/CREStereo> — веса `crestereo_eth3d.pth` (95 МБ) по
+ссылкам из его README.
+
+## Сегментация
+
+Обе модели в текущем бейзлайне **не используются** — товар отделяется от полотна
+превышением над плоскостью. Оставлены для сравнительных прогонов.
+
+* FastSAM-s.pt (23 МБ) — <https://github.com/CASIA-IVA-Lab/FastSAM>
+* yolo26n-seg.pt — <https://huggingface.co/openvision/yolo26-s-seg> (у нас вариант `n`, не `s`)
+
+## Fast FoundationStereo — проверялся, в бейзлайн не вошёл
+
+<https://github.com/NVlabs/fast-foundationstereo>
+Веса: <https://drive.google.com/drive/folders/1HuTt7UIp7gQsMiDvJwVuWmKpvFzIIMap>
+
+Брался ONNX-экспорт `23_36_37_iters_4_res_320x736` (69 МБ, id `1p9vgRh_8R1FXA79l8VdnN28dDQ3NEEhf`)
+и его yaml (`11oKt_6_jSdTqwBw1qotxmwqKxzYkvKdI`).
+
+Замер получился нечестным по времени: `onnxruntime-gpu` 1.28 требует CUDA 13, на сервере
+12.8, колёс под CUDA 12 для python 3.12 нет, `onnx2torch` падает на динамическом `Clip`,
+TensorRT не установлен — модель шла на CPU. Точность при этом от устройства не зависит и
+вышла вчетверо хуже CRE (152.8 против 32.4 мм).
+
+## Пропсы конвейера NVIDIA
+
+`assets/conveyors/` (274 МБ) тоже не в репозитории — восстанавливаются штатным
+`python3 scripts/fetch_assets.py`. Источник: `omniverse-content-production.s3-us-west-2.amazonaws.com`,
+раздел `Assets/Isaac/6.0/Isaac/Props/Conveyors/`.
+
+## Куда смотрят пути в коде
+
+Сейчас они абсолютные и указывают на сервер:
+
+* `control_test/measure_flow.py`, `control_test/cv_worker.py` — константа `CV = "/home/dasha/isaac_assets/cv"`
+* `control_test/measure_plane.py` — путь к `defom-stereo/checkpoints/`
+
+При переносе на другую машину поправить в этих трёх файлах.

assets/items/manifest.json

@@ -0,0 +1,252 @@
+{
+  "backpack": {
+    "zone": "B",
+    "gt_dims_mm": [
+      455,
+      370,
+      301
+    ],
+    "k_round": 0.82,
+    "label_ru": "Не подходит для сортировки по габаритам"
+  },
+  "bag": {
+    "zone": "D",
+    "gt_dims_mm": [
+      202,
+      175,
+      170
+    ],
+    "k_round": 0.896,
+    "label_ru": "Не подходит для сортировки без доупаковки"
+  },
+  "banana": {
+    "zone": "D",
+    "gt_dims_mm": [
+      183,
+      71,
+      33
+    ],
+    "k_round": 0.94,
+    "label_ru": "Не подходит для сортировки без доупаковки"
+  },
+  "bolts_cluster": {
+    "zone": "B",
+    "gt_dims_mm": [
+      194,
+      136,
+      53
+    ],
+    "k_round": 0.718,
+    "label_ru": "Подходит для сортировки"
+  },
+  "bottle": {
+    "zone": "D",
+    "gt_dims_mm": [
+      305,
+      91,
+      91
+    ],
+    "k_round": 0.995,
+    "label_ru": "Не подходит для сортировки без доупаковки"
+  },
+  "box_300x200x200": {
+    "zone": "B",
+    "gt_dims_mm": [
+      301,
+      200,
+      200
+    ],
+    "k_round": 0.72,
+    "label_ru": "Подходит для сортировки"
+  },
+  "box_400x400x300": {
+    "zone": "B",
+    "gt_dims_mm": [
+      401,
+      400,
+      300
+    ],
+    "k_round": 0.716,
+    "label_ru": "Не подходит для сортировки по габаритам"
+  },
+  "bucket": {
+    "zone": "D",
+    "gt_dims_mm": [
+      287,
+      287,
+      272
+    ],
+    "k_round": 0.995,
+    "label_ru": "Не подходит для сортировки без доупаковки"
+  },
+  "chip_bag": {
+    "zone": "D",
+    "gt_dims_mm": [
+      250,
+      162,
+      69
+    ],
+    "k_round": 0.811,
+    "label_ru": "Не подходит для сортировки без доупаковки"
+  },
+  "cone": {
+    "zone": "D",
+    "gt_dims_mm": [
+      500,
+      350,
+      350
+    ],
+    "k_round": 0.991,
+    "label_ru": "Не подходит для сортировки по габаритам"
+  },
+  "cylinder": {
+    "zone": "D",
+    "gt_dims_mm": [
+      435,
+      50,
+      43
+    ],
+    "k_round": 0.867,
+    "label_ru": "Не подходит для сортировки без доупаковки"
+  },
+  "detergent": {
+    "zone": "B",
+    "gt_dims_mm": [
+      278,
+      260,
+      180
+    ],
+    "k_round": 0.742,
+    "label_ru": "Подходит для сортировки"
+  },
+  "headphones": {
+    "zone": "D",
+    "gt_dims_mm": [
+      198,
+      195,
+      93
+    ],
+    "k_round": 0.807,
+    "label_ru": "Не подходит для сортировки без доупаковки"
+  },
+  "helmet": {
+    "zone": "D",
+    "gt_dims_mm": [
+      354,
+      297,
+      280
+    ],
+    "k_round": 0.895,
+    "label_ru": "Не подходит для сортировки без доупаковки"
+  },
+  "lunchbox": {
+    "zone": "B",
+    "gt_dims_mm": [
+      201,
+      152,
+      62
+    ],
+    "k_round": 0.646,
+    "label_ru": "Подходит для сортировки"
+  },
+  "mug": {
+    "zone": "D",
+    "gt_dims_mm": [
+      113,
+      99,
+      83
+    ],
+    "k_round": 0.985,
+    "label_ru": "Не подходит для сортировки без доупаковки"
+  },
+  "parcel_box": {
+    "zone": "B",
+    "gt_dims_mm": [
+      344,
+      155,
+      144
+    ],
+    "k_round": 0.699,
+    "label_ru": "Подходит для сортировки"
+  },
+  "pen": {
+    "zone": "D",
+    "gt_dims_mm": [
+      148,
+      13,
+      9
+    ],
+    "k_round": 0.842,
+    "label_ru": "Не подходит для сортировки по габаритам"
+  },
+  "perfume": {
+    "zone": "D",
+    "gt_dims_mm": [
+      120,
+      53,
+      53
+    ],
+    "k_round": 0.924,
+    "label_ru": "Не подходит для сортировки без доупаковки"
+  },
+  "pillow": {
+    "zone": "C",
+    "gt_dims_mm": [
+      455,
+      431,
+      213
+    ],
+    "k_round": 0.905,
+    "label_ru": "Не подходит для сортировки по габаритам"
+  },
+  "plate": {
+    "zone": "D",
+    "gt_dims_mm": [
+      209,
+      209,
+      27
+    ],
+    "k_round": 0.998,
+    "label_ru": "Не подходит для сортировки без доупаковки"
+  },
+  "pouf": {
+    "zone": "C",
+    "gt_dims_mm": [
+      489,
+      489,
+      264
+    ],
+    "k_round": 0.994,
+    "label_ru": "Не подходит для сортировки по габаритам"
+  },
+  "sneaker": {
+    "zone": "B",
+    "gt_dims_mm": [
+      270,
+      208,
+      125
+    ],
+    "k_round": 0.706,
+    "label_ru": "Подходит для сортировки"
+  },
+  "tool_case": {
+    "zone": "B",
+    "gt_dims_mm": [
+      300,
+      144,
+      60
+    ],
+    "k_round": 0.454,
+    "label_ru": "Подходит для сортировки"
+  },
+  "watch": {
+    "zone": "C",
+    "gt_dims_mm": [
+      230,
+      230,
+      5
+    ],
+    "k_round": 0.995,
+    "label_ru": "Не подходит для сортировки по габаритам"
+  }
+}