Reversed Mixamo "Start Climbing Ladder", retargeted to the UAL rig, then trimmed
to the step-off tail → LadderClimbDown (35f/1.13s): starts on the ladder, pushes
off, settles standing. Ships the full reversed take too (_full) as a re-trim source.
New tools:
- reverse_clip.py — pure time-reverse of a clip (30fps-pinned, sibling of pingpong_bake)
- trim_clip.py — keep a frame window, re-based to 0
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Fills Ozan order-board row #4 (avatar_pray): kneel toward statue, hands together,
hold. Mixamo Kneel held at deepest-kneel frame (184) for the legs + Praying looped
forward on the upper (rotation-only graft), retargeted to the shared Quaternius rig.
Loop QC SMOOTH (0.00deg seam). Adds tools/merge_layer.py + tools/merge_hold.py
(upper/lower track compositing: layer and hold modes).
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Female voyager bandeau top from the fc-voyager-kilt-gpt refs. Winning
construction (md_final_v16/v23): seamless-loop doctrine -- NO edge elastic
(gathering is what notched the armpit junctions), body-snug cut, zero-g wrap
(SetSimulationGravity(0) BEFORE ResetClothArrangement -- one full-gravity
materialization tick drops the bistable tube into the low basin), stiff 40f
positioning then soft 110f conform, self-measuring retry loop.
Hard-won facts logged in the scripts: strapless garments never survive full
gravity (frictionless skin + stretchy elastic); arrangement x IS azimuth
(Body_Back_Center_1 native x=0, forcing 50/50 sews both panels into a front
pouch); SetArrangementOrientation flips panel facing but mirrors it L/R;
frozen cloth renders cyan; back panel textured face points at the skin --
fix is manual Flip Normal (no API).
Texture: positioned plait/rope/fringe tile spanning the 170mm panel once
(hem anchors at v=0.5 and wraps). manual_fit_backup.zprj preserves Can's
hand-fitted drape. Exports still pending final drape approval.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
pose_clipcurl_idle.json is the recipe for extracting a clip's pinned curl;
the .txt files are the crosshand-repair vertex masks. The 230 MB of A/B bake
GLBs and render dirs stay out as regenerable scratch (now gitignored);
review/ evidence PNGs were already in the ignored tier.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
The exp05 verdict ("weights alone cannot clear the bar on this mesh") was
measured against a hard partition that was itself causing much of the tearing.
exp01-exp05 gave every vert to exactly ONE digit -- 08_finger_weights.py had an
explicit `elif lo in FING: continue # other digit: hard wall` -- which
guarantees the fused inter-digit bridges tear by the full finger separation,
because a single edge ring absorbs the whole gap.
Replace that with graded blending (WEB_BLEND_R0): a vert's fraction toward its
nearest other digit ramps 0 -> 0.5 as r = d_own/(d_own+d_other) goes 0.35 -> 0.5,
so both sides of the equidistance valley reach 50/50 and the field is continuous
across the boundary. Applied after the smoother (whose hard wall would erode beta
exactly where it must survive) and evaluated in the neighbour digit's own
arc-length frame. WEB_BLEND_SKIP excludes the thumb: its transform is opposition,
not curl, so its frame does not correspond to a finger's.
Measured against an identical baseline (same input, blend the only variable),
real tears (>=1mm rest length) drop 23-60% with NO regression on flat, the pose
that ships: fist_r 1178 -> 530, grip_r 561 -> 222, fist_l 979 -> 633,
grip_l 420 -> 322; total >5x 2487 -> 1307; p99.9 better on every pose;
flat unchanged at 0/1. fin_bones confirms the mechanism rather than just the
count -- the middle<->ring and pinky<->ring families leave the top classes while
the thumb/palm ones are untouched to the edge (182 -> 182, 148 -> 148).
This does NOT make fist/grip shippable: 222-633 real tears still reads as a
destroyed hand in clay renders, and the residual is now ~53% thumb-pad-fused-to
-palm, which is topology and needs mesh surgery or the v02 rebake. Flat and
relaxed are the shippable poses; fist/grip belong to the morph lane for now.
Also here:
- README: the solver's input is v02/..._exp03.glb, NOT exp01. exp01 is pre-hand
-fit (converter steps 2b/2c); its finger groups sit on the wrist and overlap
the real finger by 1.6cm, so a solve from it silently zeroes every _02/_03 bone
-- rigid stick fingers and a torn flat -- while weight sums stay 1.0 and every
assert passes. Cost three wasted bakes and one false "the solver regressed".
- Seed assert demanded >=100 seeds while the radius loop caps at
SEED_AXIS_R_MAX, which left pinky (88 seeds at 16mm) can never satisfy; the two
constants were mutually unsatisfiable. Now >=60, and it is documented as a
sanity gate rather than a quality bar.
- Detwist poses tested at last: real but marginal (fist_r 38.7x -> 28.7x,
grip_r 30.0x -> 17.3x, left hand flat). A knob, not a fix.
- edge_stretch_cmp.py / skin_bone_territory.py / handpose_trim_hand_obj.py:
judge tears by rest length and absolute posed growth, not raw ratio; audit
whether a bone owns any verts at all (thumb_01 owns ZERO in exp05); and trim
an arm-sized skin dump to the hand before rendering.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
The morph lane's blocking bug was believed to be coincident duplicate verts hidden
from the solver's convergence gate. Welding was necessary but nowhere near
sufficient: the previous relaxation could satisfy every stretch bar without posing
anything, so the numbers it reported were not measuring the thing they claimed.
Measured on Ariki_Female_QuatSkin_LowPoly_40, the tracked "left hand is fully
clean" baseline was a morph that moved fingertips 0.3cm, and the right hand was a
rigid 6.5cm translation of the whole hand with its shape intact. Gated Laplacian
diffusion of a delta field has a null space — constants — and edge stretch cannot
see any of it: a rigid translation stretches no edge and neither does a collapse to
zero. Both exits report perfect bars.
Six defects fixed, each with its measurement in the code comments:
weld_roi 356 duplicate groups solved twice, deltas up to 2.55cm apart
stitch_components hand is 6 overlapping sheets with 1-10mm gaps; Dijkstra
cannot cross one, so each bone saw only its own sheet
refit_fingers skeleton finger chain ran to 19.8cm; the flesh ends at 14.0cm,
so curl_02/curl_03 drove almost no weight (shipped skeleton
untouched — this is solver-local scaffolding)
solve_weights chain-arc partition of unity; the old 8mm isotropic kernels
left ZERO of 17,480 verts owned above 0.85, and a 3-way blend
averages the curl away. Plus a support cutoff: outside every
kernel, renormalized 1e-81 noise had handed a mid-palm vertex
index_02_r=0.50 and flung it 29cm
build_roi joint-sphere ROI, so the rim is a wrist band and not a fractal
of interior chart holes
relax strain-only edge projection + ROI-border seam constraints,
replacing the diffusion described above
All 8 shapes now pass every bar with real deformation: p99.9 <= 1.58x, zero edges
over 5x, zero needles with the sliver floor removed, seam <= 3.2mm, cross-hand
independence exactly 0.0000cm, mesh fingertip travel 4.6-6.2cm on fist and
3.5-4.7cm on grip (both hands). Gates now report mesh travel and seam alongside
stretch, because stretch alone cannot gate this lane.
Verified in anim_hand_test_bed: fist and grip read as a real curl on both hands,
static and mid-dance, no fins/shards/needles. It is a loose fist rather than a
clenched one — her fingers are ~4-5cm past the knuckles.
Demo GLB stays out of git (122.6 MB, ariki-game/scratchpad/). Ship decision, the
thumb-axis refit, and the TDR crash from 8 dense targets are open — see README.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
First male garment through the MD lane. MD's body analyzer rejects every
Tripo-Mako-derived mesh (0 arrangement points, 'avatar must be in T-pose or
A-pose' despite a verified 0.4-deg-perfect skeleton; same wall as ariki-game
e244e8a). Chimera test pins it on the MESH -- Lena's mesh on Mako's armature
gets 109 points. Ruled out: finger weights, decimation, head/hand chops, leg
spread, rename, OBJ import, export settings.
Workaround that ships: arrange+sew+elastic-lock the garment on a 1.07x-scaled
LENA donor, ImportFBX Mako with bAdd, DeleteAvatar(Lena), re-settle. A
seamless single-wrap slides off Mako's legs during the swap -- only SEWN
garments survive it, so the flat pattern's wrap reads are faked: overlap flap
seamed to a split front-waist segment, comb-teeth fringe (piupiu technique)
carved into the hem, rope belt as rigid Blender geometry (build_rope_belt.py).
Other findings baked into work/mako_kilt_v1/ scripts + config:
- fabric texture anchor: pattern y=0 maps to texture v=0.5 and WRAPS
(striped-probe verified; make_kilt_cloth.py mm_from_row)
- waist elastic never reaches TotalLength at default strength; strength 60
moved placement from 1.019 to 1.043
- SetAvatarActivation returns False (unusable); GetArrangementList dies on
any scene avatar mutation
QC PASS (turntable render, alpha mask, height 1.818): top 1.063 vs 1.054
target (tol .03), fringe tips 0.666 vs 0.665 (tol .04).
Avatars: _fingers = full finger weights restored via UV-space transfer from
rig-work/mako_fullhead_rigged.fbx (build_mako_avatar_fingers.py); shipped-body
avatar landed for future diagnosis. v1 (pre-flap/teeth) kept as provenance.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Marvelous Designer authoring sources for lena_hunter_skirt_v8 — the .zprj/.zpac are
hand-draped and no script regenerates them, so they are KEEP tier, not scratch — plus
the exported garment (fbx/obj/mtl), the colorway and cloth textures, and the QC
screenshots (front/back/presim) the drape was judged on.
tools/tailor/avatar/Mako_Fullhead_QuatSkin_Avatar.fbx is the MD avatar built from the
shipped male body. Worth landing even though it does not work yet: Mako's FBX imports
into MD but yields 0 arrangement points against Lena's 109, which blocks every male
garment. Having the exact file in-repo is the starting point for diagnosing that.
Mako lane intermediates (decimated fullhead, AccuRig glb, rigged blend) join their
recipes under work/mako.
All heavy files ride LFS (.glb/.blend/.fbx/.obj/.zprj/.zpac already in .gitattributes).
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
The v02 chain, which exists because bisecting the shard-eye defect proved it lives in
the hires MASTER rather than in any v02 step: the early heal chain ran whole-mesh welds
and reset custom split normals, and the eye/lash shells only read as an eye through
Tripo's authored normals. Every descendant inherits it, including the shipped body.
48_decimate_headsafe body/hands only, head bit-identical (feathered ramp)
49_seams_v02 24_seams with a density-aware hip landmark — the v01 rule fired
at u=0.610, mid-belly, on the head-protected vert count
50_seams_headuv body-only unwrap; the head KEEPS its original Tripo charts.
SLIM collapsed the undecimated lash/brow slivers to points and
ANGLE_BASED packed at half v01's texel density; the face was
the best-mapped region of the source atlas, so it is reused
51_reatlas_v02 31_reatlas + centroid splats for sub-texel triangles — the
skipped set IS the lashes, which rendered as grey glass
52_head_transplant the PRISTINE ORIGINAL head onto the decimated nude body
53_rig_transfer 54_crotch_refill
55_fullres_v02.blend is pinned in .lanekeep as THE archival master: full-res nude body
+ pristine original head, crotch refill and texture despeckle applied, 883,404 v /
1,761,640 f. It supersedes 34_v04 as the lane root (34_v04's head has the shard eyes).
Also: tools/graft_hands.py, the Marvelous Designer hunter-skirt configs v1-v8, the
hunter cloth texture generator, and Mako's measurement card.
Per .agents/rules/working-files.md the per-attempt .blend files under work/lena/v02
are SCRATCH ("never committed") — the .py recipes here are the history and regenerate
any of them from the pinned master. See the ignore rule landing next.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Ozan's hair system rebinds Quaternius styles (100% rigid to one Head bone) onto
whichever body loads, corrected by three hand-tuned constants in
CharacterBuildData.HeadFit. Those constants are only valid against the skull they
were tuned on, and nothing re-checks them when a body is rebaked. Both ship bodies
are currently wrong, in opposite directions.
tools/measure_head_shell.py:
--fit solve HeadScale/HairOffset from two cranium landmarks. A scale-about-
pivot plus a translate has exactly two degrees of freedom per axis, so
the crown + widest-cranium slice determine both outright -- no bed pass.
--sweep score every style against the clearance it was AUTHORED with on its own
stock rig (a mohawk stands proud, a beard hangs at the chin -- absolute
crown height is not the test).
default full Head-weighted shell AND the cranium band. The full shell is what
MakoHeadFit quotes; it is inflated several cm by jaw/ear/neck weight
bleed and is not the surface a cap sits on.
Measured against the shipped GLBs:
Lena cranium needs Y 0.960, carries 1.431 (+49%) -- every style floats 4.4-7.3cm
Mako cranium needs Y 1.186, carries 1.346 (+13%) -- every style sinks, buzz cut
2.2cm into his skull, moustache 8.0cm low
Solved constants take Mako to 0.00cm worst error over 10 styles, Lena to 0.29cm mean.
docs/characters/head-features-contract.md records the system, the measurements, and
what a 3-constant fit still cannot reach: Lena needs 1.613 in X but 1.311 in Z, a
23% anisotropy that stretches an authored cap sideways and drags ear cutouts along a
diagonal. That is Ozan's logged residual and wants a per-body geometry bake. Mako's
anisotropy is only 3%, so constants alone should suffice for him.
Also noted: the whole hair set shares just two base textures, so "authored per-colour
hair textures" -- the fix the wiki writes off in favour of the /0.55 swatch hack --
is 16 PNGs.
Crown placement is rest-pose arithmetic; X/Z and the in-bed look are unverified.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
The HTML references screenshots relatively, so it only renders inside a
checkout -- mailing the file on its own gives broken images. The PDF is one
self-contained file, which is what actually gets sent to Ozlem.
- tools/humans_to_pdf.py: renders .humans/*.html via headless Chrome (or
Edge), so the output is what the browser shows. HTML stays the source.
- @media print block on the page: the wide tables and the pipeline diagram
scroll on screen but would be cut off at the paper edge; this reflows
them and keeps figures, callouts and table rows off page breaks.
Verified at A4 content width -- 6 pages, both screenshots embedded.
- *.pdf -> LFS. Each regeneration is a whole new ~2 MB blob, and unlike
png/jpg there are no PDFs tracked raw yet, so there is no history to
rewrite by adding it now.
Also corrects working-files.md: it claimed .gitattributes does not LFS-track
.blend, which was true when written and is no longer -- .blend/.zprj/.obj/
.npy/.npz have since been added. The note now covers what is actually still
excluded (png/jpg, deliberately) and why that makes QA renders scratch.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Bulk import of the working lanes that were living untracked on the PC.
Content:
- characters/ Lena/male body lanes, bakes, texture work, run logs
- clothing/ garment pipeline, configs, gates, contract docs
- garments/ MD-authored garment sources (.zprj/.zpac)
- UAL-Lib/ Universal Animation Library 2 source (.blend/.fbx/.glb)
- tools/ blender_bridge, iclone_bridge, md_bridge, tailor, glm_agent
- docs/, plans/, dev/, .agents/plans/
Repo hygiene:
- .gitattributes: LFS now covers .blend, .zprj, .zpac, .obj, .npy and the
Reallusion .iAvatar/.ccAvatar/.ccRestore containers. Without this the
~3.8 GB in this commit would land as raw blobs. .png/.jpg are left out
on purpose — ~250 are already tracked raw and converting them would
rewrite every one without shrinking history.
- .gitignore: exclude /accurig/ (~1 GB AccuRig program files, redistributable
from Reallusion, nothing authored here) and /dev/null/ (git-lfs hook copies
dropped by a `>/dev/null` redirect on Windows).
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Two pieces of work.
1. Keep milestones, not steps. Staged NN_*.py lanes were saving a full
~80 MB .blend per attempt, so Lena's lane reached 3.0 GB of which 2.2 GB
was 30 .blend files -- five snapshots to land one crotch fix, five more
for the bra. The .py recipes are the real history; the blends are cache.
- .agents/rules/working-files.md: four-tier policy (KEEP / MASTER /
SCRATCH / UNKNOWN) + how to work a lane.
- tools/prune_lane.py: classifies a lane and prunes the scratch tier.
Dry-run by default. Reads a per-lane .lanekeep manifest, flags
binaries byte-identical to a registered original (canonical name
always survives a duplicate pair), and never auto-deletes a .blend
with no step script beside it -- those cannot be rebuilt.
- .gitignore: scratch patterns can never be committed.
Dry run on characters/female/lena_nude reports 2.3 GB reclaimable.
Not applied -- that lane had a live Blender session at the time.
2. .humans/marvelous-designer.html: how we author garments in Marvelous
Designer, written for people rather than agents -- the six-step process,
what has been made, the traps that cost hours, and what is still
unsolved. Matches the .humans/ HTML convention in ariki-game.
Also committing the docs the AGENTS.md knowledge map and the new page
reference, so they are not dangling: the clothing-lane architecture page,
the marvelous-designer skill, and the two screenshots the page embeds.
characters/ is deliberately untracked and stays that way -- it holds GBs of
blends and GLBs, and .gitattributes does not LFS-track .blend.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
From the conformance audit (ariki-game .agents/audits/animation-standard-assessment-2026-07-23.md):
- tools/: cc_retarget + mixamo_retarget re-synced from ariki-game (were stale forks
missing --despike/--fingers/--map ual); kevin_retarget path now repo-relative;
mirror rule documented in exchange/GUIDE.md
- INBOX/lena: add canonical Ariki_Female_QuatSkin.glb; README marks MixamoSkin superseded
- REGISTRY: tide_dance_01 flagged (no DanceType enum entry); rain_dance note corrected
(RainDance IS in the enum); boat table notes live canoedismount1/run_to_dive; QC split
(loop_qc pre-commit vs anim_qc post-ingest) documented
- docs/iclone-bridge.md: stub created (GUIDE referenced it but it never existed)
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
tools/dance_profile.py samples world-space bone positions per frame in
Blender headless and emits a per-beat motion-energy timeline (accents,
phrases, totals) at a given BPM. Profiles generated for war_dance_01,
fertility_dance_01, taming_dance_01.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
tools/export_boat_prop.py (GLM-implemented, 2 rounds): hull at game scale
(8.5 m, keel 0, no recenter) + placeholder mast/sail/steering oar + 5 seat
markers + waterline + 1.9 m ref figure; self-verify incl. right-side-up
gates added after round 1 shipped an upside-down hull with all checks
green. Registry: boat actions table (6 slots, direct names). README
carries CC BY 4.0 attribution + iClone import/filming notes.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Keep aloha1(+pp), dance1(+pp), hakadance1. Archive hakadance2, alohaOG,
firedance1, hakaOG (all variants) to archive/. New dances get provisional
nd## ids at conversion; rename to a dance code (haka01 etc.) only on
adoption via tools/rename_clip.py. Registry: docs/dances/REGISTRY.md.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Forward+reversed baked as one clip: end==start by construction, drift
self-cancels, full choreography kept; motion reverses at the turn point.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Implemented by GLM from plans/loop-fix-plan-2026-07-16.md; all seven clips
verified SMOOTH by loop_qc (0° pose gap, 0 cm drift, 60-89% duration kept).
Results: plans/loop-fix-results-2026-07-16.md.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Pose gap / velocity gap / root drift at the LoopMode.Linear wrap; exit-code
gated. Skill updated: loop QC is workflow step 5, with the Blender-side
fixer playbook.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
jeremy
Can Narin