Add verified DiffuseStyleGesture+ and joint-only uTMR

.gitattributes

@@ -111,3 +111,18 @@ cases/assets/2_scott_0_81_81_emage.smpl filter=lfs diff=lfs merge=lfs -text
 cases/assets/2_scott_0_87_87_emage.smpl filter=lfs diff=lfs merge=lfs -text
 cases/assets/2_scott_0_8_8_emage.smpl filter=lfs diff=lfs merge=lfs -text
 cases/assets/2_scott_0_95_95_emage.smpl filter=lfs diff=lfs merge=lfs -text
+cases/assets/2_scott_0_103_103_diffusestylegesture.smpl filter=lfs diff=lfs merge=lfs -text
+cases/assets/2_scott_0_111_111_diffusestylegesture.smpl filter=lfs diff=lfs merge=lfs -text
+cases/assets/2_scott_0_1_1_diffusestylegesture.smpl filter=lfs diff=lfs merge=lfs -text
+cases/assets/2_scott_0_2_2_diffusestylegesture.smpl filter=lfs diff=lfs merge=lfs -text
+cases/assets/2_scott_0_3_3_diffusestylegesture.smpl filter=lfs diff=lfs merge=lfs -text
+cases/assets/2_scott_0_4_4_diffusestylegesture.smpl filter=lfs diff=lfs merge=lfs -text
+cases/assets/2_scott_0_5_5_diffusestylegesture.smpl filter=lfs diff=lfs merge=lfs -text
+cases/assets/2_scott_0_65_65_diffusestylegesture.smpl filter=lfs diff=lfs merge=lfs -text
+cases/assets/2_scott_0_6_6_diffusestylegesture.smpl filter=lfs diff=lfs merge=lfs -text
+cases/assets/2_scott_0_73_73_diffusestylegesture.smpl filter=lfs diff=lfs merge=lfs -text
+cases/assets/2_scott_0_7_7_diffusestylegesture.smpl filter=lfs diff=lfs merge=lfs -text
+cases/assets/2_scott_0_81_81_diffusestylegesture.smpl filter=lfs diff=lfs merge=lfs -text
+cases/assets/2_scott_0_87_87_diffusestylegesture.smpl filter=lfs diff=lfs merge=lfs -text
+cases/assets/2_scott_0_8_8_diffusestylegesture.smpl filter=lfs diff=lfs merge=lfs -text
+cases/assets/2_scott_0_95_95_diffusestylegesture.smpl filter=lfs diff=lfs merge=lfs -text

cases/assets/2_scott_0_103_103_diffusestylegesture.smpl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4b9c881414c470fd35442bdfcd68354298e8114bde1827655e78f5284be287c3
+size 459810

cases/assets/2_scott_0_111_111_diffusestylegesture.smpl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4938fee55b5268cca45d878420d03123634f7e21fba0210efe77b137121e8752
+size 484380

cases/assets/2_scott_0_1_1_diffusestylegesture.smpl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:67dd5e22ac03cca603e77fc2c66412c42cd9627edffab1611fc73052e7ded0d6
+size 514620

cases/assets/2_scott_0_2_2_diffusestylegesture.smpl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:15e4e8d964b7f8b15d4bc80485b36d5d5a3ab446c66d13b1039a4f857989a1f0
+size 500580

cases/assets/2_scott_0_3_3_diffusestylegesture.smpl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4eff364260a3e65343abb547f2d103c2ff8055c4fa49b7b1eb9093fb85253177
+size 549180

cases/assets/2_scott_0_4_4_diffusestylegesture.smpl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1be34c7dd8cafdb837cb715476ef900e41e0304db82e5f1f7235052c6144a228
+size 541080

cases/assets/2_scott_0_5_5_diffusestylegesture.smpl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a8401d4fa1c47ee94cebd3134d02ba3e22b670c7b186c594f121582b3b44a1bb
+size 622080

cases/assets/2_scott_0_65_65_diffusestylegesture.smpl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a9bc504eb0a730edcfbefa05b1a8a5ae3267a8397873457bd5164b461ab393db
+size 427680

cases/assets/2_scott_0_6_6_diffusestylegesture.smpl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e1266d9f7e1a24560079f92a01ee34556c6e8c0d3330956df20e71af62d6c7ea
+size 670680

cases/assets/2_scott_0_73_73_diffusestylegesture.smpl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a644b3592cbd1960b784f26ed7fb93b1369e10b8ad9fdbc3830992be5af74b51
+size 476280

cases/assets/2_scott_0_7_7_diffusestylegesture.smpl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cf1ef9c6e47f6b0d553deb1033734da0546f39044f4e549d37da9dc425193735
+size 565380

cases/assets/2_scott_0_81_81_diffusestylegesture.smpl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8361f23ac7eb4631930ae50a2815dc1a1aa5767d1be0d0eaade4e4c1799594d0
+size 540810

cases/assets/2_scott_0_87_87_diffusestylegesture.smpl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:72eca7de2f04aed2120c013232616c5c0b55c43f91fdca95d28617dc15e8f900
+size 630180

cases/assets/2_scott_0_8_8_diffusestylegesture.smpl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ede5589ccce97f8d2f6d6e576d8bf251b71f8212b6a78a74fa7234f5db809548
+size 628290

cases/assets/2_scott_0_95_95_diffusestylegesture.smpl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:528cb2c796d836db272d502d03c5ac04253fbcf9c4a847f8630e80094dfcf214
+size 395280

cases/manifest.json

@@ -23,15 +23,20 @@
       "key": "emage",
       "label": "EMAGE",
       "accent": "#315f9d"
+    },
+    {
+      "key": "diffusestylegesture",
+      "label": "DiffuseStyleGesture+",
+      "accent": "#7b3fa1"
     }
   ],
   "checkpoint_scope": "public_demo_visualization_only",
   "provenance": {
     "dataset": "BEAT2 English v2.0.0",
     "speaker": 2,
-    "artifact": "ZeyuLing/Motius-Language-of-Motion",
+    "artifact": "JuzeZhang/language_of_motion",
     "seed_policy": "reset_official_seed_before_each_sample",
-    "visualization": "Shared neutral SMPL-22 body projection of the native SMPL-H/SMPL-X poses; LoM and EMAGE fingers and expressions are retained in raw/ but not rendered by this web mesh."
+    "visualization": "Shared neutral SMPL-22 body projection of the native SMPL-H/SMPL-X poses; LoM, EMAGE, and DiffuseStyleGesture+ fingers and expressions are retained in raw/ but not rendered by this web mesh."
   },
   "cases": [
     {
@@ -104,6 +109,27 @@
           "translation_offset": 0,
           "rotation_offset": 10218,
           "fps": 30.0
+        },
+        "diffusestylegesture": {
+          "frames": 1703,
+          "display_frames": 1703,
+          "stride": 1,
+          "translation_count": 5109,
+          "rotation_count": 224796,
+          "translation_minimum": [
+            -0.287344753742218,
+            -1.3640080690383911,
+            -0.11482555419206619
+          ],
+          "translation_scale": [
+            9.071621207112912e-06,
+            3.5692999517777935e-05,
+            3.4177699035353726e-06
+          ],
+          "asset": "assets/2_scott_0_103_103_diffusestylegesture.smpl",
+          "translation_offset": 0,
+          "rotation_offset": 10218,
+          "fps": 30.0
         }
       }
     },
@@ -177,6 +203,27 @@
           "translation_offset": 0,
           "rotation_offset": 10764,
           "fps": 30.0
+        },
+        "diffusestylegesture": {
+          "frames": 1794,
+          "display_frames": 1794,
+          "stride": 1,
+          "translation_count": 5382,
+          "rotation_count": 236808,
+          "translation_minimum": [
+            -0.3729633092880249,
+            -0.7559877634048462,
+            -0.10018095374107361
+          ],
+          "translation_scale": [
+            7.0251940087473486e-06,
+            1.4079354514251463e-05,
+            1.9060267959503108e-06
+          ],
+          "asset": "assets/2_scott_0_111_111_diffusestylegesture.smpl",
+          "translation_offset": 0,
+          "rotation_offset": 10764,
+          "fps": 30.0
         }
       }
     },
@@ -250,6 +297,27 @@
           "translation_offset": 0,
           "rotation_offset": 11436,
           "fps": 30.0
+        },
+        "diffusestylegesture": {
+          "frames": 1906,
+          "display_frames": 1906,
+          "stride": 1,
+          "translation_count": 5718,
+          "rotation_count": 251592,
+          "translation_minimum": [
+            -0.10285009443759918,
+            -0.37792709469795227,
+            -0.01848311722278595
+          ],
+          "translation_scale": [
+            3.961683887609979e-06,
+            8.289023753604852e-06,
+            7.397749186566216e-07
+          ],
+          "asset": "assets/2_scott_0_1_1_diffusestylegesture.smpl",
+          "translation_offset": 0,
+          "rotation_offset": 11436,
+          "fps": 30.0
         }
       }
     },
@@ -323,6 +391,27 @@
           "translation_offset": 0,
           "rotation_offset": 11124,
           "fps": 30.0
+        },
+        "diffusestylegesture": {
+          "frames": 1854,
+          "display_frames": 1854,
+          "stride": 1,
+          "translation_count": 5562,
+          "rotation_count": 244728,
+          "translation_minimum": [
+            -0.1015571653842926,
+            -0.3336181342601776,
+            -0.02136019617319107
+          ],
+          "translation_scale": [
+            4.766444362758193e-06,
+            6.343251698126551e-06,
+            8.156304147632909e-07
+          ],
+          "asset": "assets/2_scott_0_2_2_diffusestylegesture.smpl",
+          "translation_offset": 0,
+          "rotation_offset": 11124,
+          "fps": 30.0
         }
       }
     },
@@ -396,6 +485,27 @@
           "translation_offset": 0,
           "rotation_offset": 12204,
           "fps": 30.0
+        },
+        "diffusestylegesture": {
+          "frames": 2034,
+          "display_frames": 2034,
+          "stride": 1,
+          "translation_count": 6102,
+          "rotation_count": 268488,
+          "translation_minimum": [
+            -0.1521918922662735,
+            -0.41209226846694946,
+            -0.07307065278291702
+          ],
+          "translation_scale": [
+            4.904836259811418e-06,
+            8.358090781257488e-06,
+            1.491649413765117e-06
+          ],
+          "asset": "assets/2_scott_0_3_3_diffusestylegesture.smpl",
+          "translation_offset": 0,
+          "rotation_offset": 12204,
+          "fps": 30.0
         }
       }
     },
@@ -469,6 +579,27 @@
           "translation_offset": 0,
           "rotation_offset": 12024,
           "fps": 30.0
+        },
+        "diffusestylegesture": {
+          "frames": 2004,
+          "display_frames": 2004,
+          "stride": 1,
+          "translation_count": 6012,
+          "rotation_count": 264528,
+          "translation_minimum": [
+            -0.2579508423805237,
+            -0.2772132456302643,
+            -0.0752326101064682
+          ],
+          "translation_scale": [
+            6.928152743057581e-06,
+            8.001823516678996e-06,
+            1.8671813677428872e-06
+          ],
+          "asset": "assets/2_scott_0_4_4_diffusestylegesture.smpl",
+          "translation_offset": 0,
+          "rotation_offset": 12024,
+          "fps": 30.0
         }
       }
     },
@@ -542,6 +673,27 @@
           "translation_offset": 0,
           "rotation_offset": 13824,
           "fps": 30.0
+        },
+        "diffusestylegesture": {
+          "frames": 2304,
+          "display_frames": 2304,
+          "stride": 1,
+          "translation_count": 6912,
+          "rotation_count": 304128,
+          "translation_minimum": [
+            -0.1123095378279686,
+            -0.47919392585754395,
+            -0.023244939744472504
+          ],
+          "translation_scale": [
+            6.45385716779856e-06,
+            9.558820238453336e-06,
+            1.2164429108452168e-06
+          ],
+          "asset": "assets/2_scott_0_5_5_diffusestylegesture.smpl",
+          "translation_offset": 0,
+          "rotation_offset": 13824,
+          "fps": 30.0
         }
       }
     },
@@ -615,6 +767,27 @@
           "translation_offset": 0,
           "rotation_offset": 9504,
           "fps": 30.0
+        },
+        "diffusestylegesture": {
+          "frames": 1584,
+          "display_frames": 1584,
+          "stride": 1,
+          "translation_count": 4752,
+          "rotation_count": 209088,
+          "translation_minimum": [
+            -0.497951865196228,
+            -0.3282739222049713,
+            -0.05646343156695366
+          ],
+          "translation_scale": [
+            1.3638137716043275e-05,
+            1.4359416127263103e-05,
+            1.2987100035388721e-06
+          ],
+          "asset": "assets/2_scott_0_65_65_diffusestylegesture.smpl",
+          "translation_offset": 0,
+          "rotation_offset": 9504,
+          "fps": 30.0
         }
       }
     },
@@ -688,6 +861,27 @@
           "translation_offset": 0,
           "rotation_offset": 14904,
           "fps": 30.0
+        },
+        "diffusestylegesture": {
+          "frames": 2484,
+          "display_frames": 2484,
+          "stride": 1,
+          "translation_count": 7452,
+          "rotation_count": 327888,
+          "translation_minimum": [
+            -0.14530745148658752,
+            -0.37546154856681824,
+            -0.02249929867684841
+          ],
+          "translation_scale": [
+            5.042185875936411e-06,
+            9.032469279190991e-06,
+            8.246228730968141e-07
+          ],
+          "asset": "assets/2_scott_0_6_6_diffusestylegesture.smpl",
+          "translation_offset": 0,
+          "rotation_offset": 14904,
+          "fps": 30.0
         }
       }
     },
@@ -761,6 +955,27 @@
           "translation_offset": 0,
           "rotation_offset": 10584,
           "fps": 30.0
+        },
+        "diffusestylegesture": {
+          "frames": 1764,
+          "display_frames": 1764,
+          "stride": 1,
+          "translation_count": 5292,
+          "rotation_count": 232848,
+          "translation_minimum": [
+            -0.2553229033946991,
+            -1.0471454858779907,
+            -0.052459314465522766
+          ],
+          "translation_scale": [
+            1.1358723895682488e-05,
+            2.7841984774568118e-05,
+            2.0742697870446136e-06
+          ],
+          "asset": "assets/2_scott_0_73_73_diffusestylegesture.smpl",
+          "translation_offset": 0,
+          "rotation_offset": 10584,
+          "fps": 30.0
         }
       }
     },
@@ -834,6 +1049,27 @@
           "translation_offset": 0,
           "rotation_offset": 12564,
           "fps": 30.0
+        },
+        "diffusestylegesture": {
+          "frames": 2094,
+          "display_frames": 2094,
+          "stride": 1,
+          "translation_count": 6282,
+          "rotation_count": 276408,
+          "translation_minimum": [
+            -0.2046271413564682,
+            -0.3428315222263336,
+            -0.0661189928650856
+          ],
+          "translation_scale": [
+            7.567646207462531e-06,
+            7.513385753554758e-06,
+            2.126978188243811e-06
+          ],
+          "asset": "assets/2_scott_0_7_7_diffusestylegesture.smpl",
+          "translation_offset": 0,
+          "rotation_offset": 12564,
+          "fps": 30.0
         }
       }
     },
@@ -907,6 +1143,27 @@
           "translation_offset": 0,
           "rotation_offset": 12018,
           "fps": 30.0
+        },
+        "diffusestylegesture": {
+          "frames": 2003,
+          "display_frames": 2003,
+          "stride": 1,
+          "translation_count": 6009,
+          "rotation_count": 264396,
+          "translation_minimum": [
+            -0.25451359152793884,
+            -0.36991220712661743,
+            -0.04123775288462639
+          ],
+          "translation_scale": [
+            5.130808858666569e-06,
+            7.841794285923243e-06,
+            8.533477853234217e-07
+          ],
+          "asset": "assets/2_scott_0_81_81_diffusestylegesture.smpl",
+          "translation_offset": 0,
+          "rotation_offset": 12018,
+          "fps": 30.0
         }
       }
     },
@@ -980,6 +1237,27 @@
           "translation_offset": 0,
           "rotation_offset": 14004,
           "fps": 30.0
+        },
+        "diffusestylegesture": {
+          "frames": 2334,
+          "display_frames": 2334,
+          "stride": 1,
+          "translation_count": 7002,
+          "rotation_count": 308088,
+          "translation_minimum": [
+            -0.33953434228897095,
+            -0.897826611995697,
+            -0.05458662658929825
+          ],
+          "translation_scale": [
+            7.330309927056078e-06,
+            1.3669889995071571e-05,
+            1.4377337720361538e-06
+          ],
+          "asset": "assets/2_scott_0_87_87_diffusestylegesture.smpl",
+          "translation_offset": 0,
+          "rotation_offset": 14004,
+          "fps": 30.0
         }
       }
     },
@@ -1053,6 +1331,27 @@
           "translation_offset": 0,
           "rotation_offset": 13962,
           "fps": 30.0
+        },
+        "diffusestylegesture": {
+          "frames": 2327,
+          "display_frames": 2327,
+          "stride": 1,
+          "translation_count": 6981,
+          "rotation_count": 307164,
+          "translation_minimum": [
+            -0.09713488072156906,
+            -0.35423916578292847,
+            -0.01455355528742075
+          ],
+          "translation_scale": [
+            5.563640115724411e-06,
+            8.990169590106234e-06,
+            6.951847808522871e-07
+          ],
+          "asset": "assets/2_scott_0_8_8_diffusestylegesture.smpl",
+          "translation_offset": 0,
+          "rotation_offset": 13962,
+          "fps": 30.0
         }
       }
     },
@@ -1126,6 +1425,27 @@
           "translation_offset": 0,
           "rotation_offset": 8784,
           "fps": 30.0
+        },
+        "diffusestylegesture": {
+          "frames": 1464,
+          "display_frames": 1464,
+          "stride": 1,
+          "translation_count": 4392,
+          "rotation_count": 193248,
+          "translation_minimum": [
+            -0.35380634665489197,
+            -0.6937940120697021,
+            -0.15583853423595428
+          ],
+          "translation_scale": [
+            7.507933332817629e-06,
+            1.4125566849543247e-05,
+            2.7392625270294957e-06
+          ],
+          "asset": "assets/2_scott_0_95_95_diffusestylegesture.smpl",
+          "translation_offset": 0,
+          "rotation_offset": 8784,
+          "fps": 30.0
         }
       }
     }

cases/raw/diffusestylegesture/2_scott_0_103_103.npz

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3e9bafe23e54859f9ab39c019f76efa7ac0a4c6f1c7eb3b06cd63f69e097d959
+size 1039587

cases/raw/diffusestylegesture/2_scott_0_111_111.npz

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ea8090f38ee4c279b1590938e27071951090e2fda34c813fe4bf60e68fb1316c
+size 1093989

cases/raw/diffusestylegesture/2_scott_0_1_1.npz

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:53cc656048909b8151d7fc69fd4953d0b297826db97e0dff2664a6a9879c7073
+size 1161176

cases/raw/diffusestylegesture/2_scott_0_2_2.npz

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:aaa6532ecf6cd526479cda98f9b999e3f3ef8877402e2dc434c74255986de15f
+size 1130356

cases/raw/diffusestylegesture/2_scott_0_3_3.npz

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b2b91becbd32c61ffab1100bafadba2972413d129928c4717906cfff2cf381d6
+size 1240255

cases/raw/diffusestylegesture/2_scott_0_4_4.npz

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1f37717a2e08dcfb07e6e4ad8512f1a26f013e7992a4105dd00893d806bba69b
+size 1222825

cases/raw/diffusestylegesture/2_scott_0_5_5.npz

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:95e01ab61404527aac4f7c3ed9fe0500ac7e29de586e808c0eabb5a46dc5e71c
+size 1404455

cases/raw/diffusestylegesture/2_scott_0_65_65.npz

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9027997df30a52ff2210451146e6b8ed0452ea5652373c3033b194337080194b
+size 965962

cases/raw/diffusestylegesture/2_scott_0_6_6.npz

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9ff9f58f2e81350c780c4023d2e83642b4ad8cac0428f0cc66290498e928bf25
+size 1512745

cases/raw/diffusestylegesture/2_scott_0_73_73.npz

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:458ba7f473b69b843b462d2fcd96cf2097401da0296c27f1ed864e12563060a4
+size 1075898

cases/raw/diffusestylegesture/2_scott_0_7_7.npz

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6e6f598fe88940fbbc3defeb3b25f81bab6fbea1cf3afafda494590a42a8cd9c
+size 1276491

cases/raw/diffusestylegesture/2_scott_0_81_81.npz

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0e4048d69ab5bfa0616bc5c35e4a1b54c1cc0b10396b49b97affd0ab2716f11c
+size 1219927

cases/raw/diffusestylegesture/2_scott_0_87_87.npz

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:14955460ab3abcdcca083b58a9248ec4b4db5fb3afa68090143bd263a3ca98b3
+size 1422682

cases/raw/diffusestylegesture/2_scott_0_8_8.npz

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:379f6ca3113e6d68804434d9ec00c4e0d0f219e40d7e58ae16e078cc22c5ee45
+size 1418068

cases/raw/diffusestylegesture/2_scott_0_95_95.npz

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0b8ef587a9817b85d36c986686e07250506ec1f808cd3220c805d4b773507d89
+size 893574

index.html

@@ -13,8 +13,8 @@
     .topbar{display:flex;align-items:flex-start;justify-content:space-between;gap:20px;margin-bottom:18px;border-bottom:1px solid var(--line);padding-bottom:22px}h1{margin:0;font-size:30px;line-height:1.15}.subtitle{max-width:880px;margin:10px 0 0;color:var(--muted);font-size:14px}.badge-row{display:flex;flex-wrap:wrap;gap:8px;margin-top:14px}.badge{display:inline-flex;align-items:center;border:1px solid var(--line);background:var(--panel);border-radius:999px;padding:5px 10px;color:#3d4857;font-size:12px;white-space:nowrap}.updated{flex:0 0 auto;min-width:184px;border:1px solid var(--line);background:var(--panel);border-radius:8px;padding:10px 12px;color:var(--muted);font-size:12px;box-shadow:var(--shadow)}.updated strong{display:block;color:var(--ink);font-size:14px;font-weight:650}
     .score-strip{display:grid;grid-template-columns:repeat(4,minmax(0,1fr));gap:12px;margin:18px 0}.score-card{border:1px solid var(--line);border-top:3px solid var(--teal);border-radius:8px;background:var(--panel);padding:14px;box-shadow:var(--shadow)}.score-card:nth-child(2){border-top-color:var(--blue)}.score-card:nth-child(3){border-top-color:var(--amber)}.score-card:nth-child(4){border-top-color:var(--coral)}.score-card .label{color:var(--muted);font-size:12px;text-transform:uppercase}.score-card .value{min-height:31px;margin-top:6px;font-size:21px;font-weight:720}.score-card .note{margin-top:4px;color:var(--muted);font-size:12px}
     section.panel{margin:18px 0;border:1px solid var(--line);border-radius:8px;background:var(--panel);box-shadow:var(--shadow);overflow:hidden}.section-heading{display:flex;align-items:flex-end;justify-content:space-between;gap:20px;border-bottom:1px solid var(--line);padding:16px 18px}.section-heading h2{margin:0;font-size:18px}.section-heading p{margin:3px 0 0;color:var(--muted);font-size:12px}.section-link{display:inline-flex;align-items:center;min-height:34px;border:1px solid var(--line-strong);border-radius:6px;padding:0 11px;background:#fff;color:var(--teal);font-size:12px;font-weight:650;text-decoration:none}.section-link:hover{border-color:var(--teal)}
-    .explorer-frame{display:block;width:100%;height:900px;border:0;background:#eef2f0}.table-wrap{overflow:auto}.leaderboard{width:100%;min-width:1110px;border-collapse:separate;border-spacing:0;font-size:13px}.leaderboard th,.leaderboard td{padding:11px 12px;border-bottom:1px solid var(--line);text-align:right;white-space:nowrap}.leaderboard thead th{background:#f2f5f8;color:#425064;font-size:11px}.leaderboard thead tr:first-child th{border-bottom:1px solid var(--line-strong);text-align:center}.leaderboard th:first-child,.leaderboard td:first-child{text-align:left}.leaderboard tbody tr:hover{background:#f6f9f7}.leaderboard tbody tr.reference{background:#eef3f1}.method strong{display:block}.method small{color:var(--muted);font-size:11px}.rank{display:inline-grid;place-items:center;width:20px;height:20px;margin-right:8px;border-radius:50%;background:#e8eeeb;color:#46524d;font:11px "IBM Plex Mono",monospace}.status{display:inline-flex;padding:4px 7px;border:1px solid var(--line);border-radius:4px;font:9px "IBM Plex Mono",monospace;text-transform:uppercase}.status.verified{border-color:#74a995;background:#e4f0e8;color:#145b47}.best{font-weight:800;color:#086b62}.note{padding:13px 18px;color:var(--muted);font-size:12px}
-    .protocol{display:grid;grid-template-columns:repeat(3,minmax(0,1fr))}.protocol article{padding:18px;border-right:1px solid var(--line)}.protocol article:last-child{border:0}.protocol h3{margin:0 0 6px;font-size:13px}.protocol p{margin:0;color:var(--muted);font-size:12px}.protocol code{font-family:"IBM Plex Mono",monospace;font-size:11px;color:var(--teal)}
+    .explorer-frame{display:block;width:100%;height:900px;border:0;background:#eef2f0}.table-wrap{overflow:auto}.leaderboard{width:100%;min-width:1360px;border-collapse:separate;border-spacing:0;font-size:13px}.leaderboard th,.leaderboard td{padding:11px 12px;border-bottom:1px solid var(--line);text-align:right;white-space:nowrap}.leaderboard thead th{background:#f2f5f8;color:#425064;font-size:11px}.leaderboard thead tr:first-child th{border-bottom:1px solid var(--line-strong);text-align:center}.leaderboard th:first-child,.leaderboard td:first-child{text-align:left}.leaderboard tbody tr:hover{background:#f6f9f7}.leaderboard tbody tr.reference{background:#eef3f1}.method strong{display:block}.method small{color:var(--muted);font-size:11px}.rank{display:inline-grid;place-items:center;width:20px;height:20px;margin-right:8px;border-radius:50%;background:#e8eeeb;color:#46524d;font:11px "IBM Plex Mono",monospace}.status{display:inline-flex;padding:4px 7px;border:1px solid var(--line);border-radius:4px;font:9px "IBM Plex Mono",monospace;text-transform:uppercase}.status.verified{border-color:#74a995;background:#e4f0e8;color:#145b47}.best{font-weight:800;color:#086b62}.note{padding:13px 18px;color:var(--muted);font-size:12px}
+    .protocol{display:grid;grid-template-columns:repeat(4,minmax(0,1fr))}.protocol article{padding:18px;border-right:1px solid var(--line)}.protocol article:last-child{border:0}.protocol h3{margin:0 0 6px;font-size:13px}.protocol p{margin:0;color:var(--muted);font-size:12px}.protocol code{font-family:"IBM Plex Mono",monospace;font-size:11px;color:var(--teal)}
     .parity-wrap{padding:0 18px 18px}.parity{width:100%;border-collapse:collapse}.parity th,.parity td{padding:10px;border-bottom:1px solid var(--line);text-align:right}.parity th:first-child,.parity td:first-child{text-align:left}.parity td:first-child strong,.parity td:first-child span{display:block}.parity td:first-child span{color:var(--muted);font-size:11px}.parity tr.delta{background:#f6f9f7;color:var(--muted)}
     .footer{display:flex;justify-content:space-between;gap:20px;padding:8px 0 0;color:var(--muted);font-size:12px}
     @media(max-width:980px){.topbar{flex-direction:column}.updated{width:100%}.score-strip{grid-template-columns:repeat(2,1fr)}.protocol{grid-template-columns:1fr}.protocol article{border-right:0;border-bottom:1px solid var(--line)}.explorer-frame{height:820px}}
@@ -56,7 +56,7 @@
     </section>
 
     <section class="panel" id="case-explorer">
-      <div class="section-heading"><div><h2>Interactive Three.js Motion Comparison</h2><p>High-resolution, camera-controllable GT, Language of Motion, and EMAGE previews across all 15 clips. Full native SMPL-X predictions remain downloadable as NPZ.</p></div><a class="section-link" href="cases/index.html" target="_blank" rel="noopener noreferrer">Open full screen</a></div>
+      <div class="section-heading"><div><h2>Interactive Three.js Motion Comparison</h2><p>High-resolution, camera-controllable GT, Language of Motion, EMAGE, and DiffuseStyleGesture+ previews across all 15 clips. Full native SMPL-X predictions remain downloadable as NPZ.</p></div><a class="section-link" href="cases/index.html" target="_blank" rel="noopener noreferrer">Open full screen</a></div>
       <iframe class="explorer-frame" src="cases/index.html" title="BEAT2 audio-synchronized all-case 3D motion comparison" loading="lazy" allow="autoplay"></iframe>
     </section>
 
@@ -64,11 +64,11 @@
       <div class="section-heading"><div><h2>Results</h2><p>Lower is better for FGD and paired errors; higher is better for BC and Diversity.</p></div></div>
       <div class="table-wrap">
         <table class="leaderboard">
-          <thead><tr><th rowspan="2">Method</th><th rowspan="2">Status</th><th colspan="3">Official BEAT2 metrics</th><th colspan="3">Paired diagnostics</th></tr><tr><th>FGD ↓</th><th>BC ↑</th><th>Diversity ↑</th><th>Rotation ↓</th><th>Expression ↓</th><th>Translation ↓</th></tr></thead>
+          <thead><tr><th rowspan="2">Method</th><th rowspan="2">Status</th><th colspan="3">Official BEAT2 metrics</th><th colspan="2">Joint-only uTMR</th><th colspan="3">Paired diagnostics</th></tr><tr><th>FGD ↓</th><th>BC ↑</th><th>Diversity ↑</th><th>FID ↓</th><th>Paired Dist. ↓</th><th>Rotation ↓</th><th>Expression ↓</th><th>Translation ↓</th></tr></thead>
           <tbody id="rows"></tbody>
         </table>
       </div>
-      <div class="note">FGD and BC use the paper display scale (raw evaluator value ×10). The public LoM demo checkpoint is verified but is not labeled as the paper's private/full evaluation checkpoint.</div>
+      <div class="note">FGD and BC use the paper display scale (raw evaluator value ×10). uTMR consumes only canonical SMPL-22 joints and L2-normalized motion embeddings; no transcript, expression, or raw root translation enters those columns.</div>
     </section>
 
     <section class="panel" id="protocol">
@@ -77,6 +77,7 @@
         <article><h3>Fréchet Gesture Distance</h3><p>Released <code>AESKConv_240_100</code> features over 32-frame-aligned SMPL-X rotations with the official covariance regularization.</p></article>
         <article><h3>Beat Correlation</h3><p>Audio onsets against upper-body velocity minima, excluding the official 60-frame margin at each end.</p></article>
         <article><h3>L1 Diversity</h3><p>Mean absolute deviation of generated 55-joint positions after the official SMPL-X metric forward kinematics.</p></article>
+        <article><h3>Joint-only uTMR</h3><p>Universal SMPL-H joints66 motion embeddings over balanced ≤20-second windows, with per-sample L2 normalization for FID and paired distance.</p></article>
       </div>
     </section>
 

leaderboard.js

@@ -1,6 +1,7 @@
 const fmt=(value,digits=4)=>value==null||!Number.isFinite(Number(value))?"—":Number(value).toFixed(digits);
 const metricSpecs=[
   {key:"fgd",lower:true},{key:"bc",lower:false},{key:"diversity",lower:false},
+  {key:"utmr_fid",lower:true},{key:"utmr_paired_distance",lower:true},
   {key:"rotation_geodesic",lower:true},{key:"expression_l2",lower:true},{key:"translation_l2",lower:true}
 ];
 
@@ -57,7 +58,7 @@ function render(data){
   document.getElementById("rows").innerHTML=data.rows.map(row=>{
     const primary=ranks.fgd.get(row.method);
     const marker=row.kind==="reference"?"GT":primary||"·";
-    return `<tr class="${row.kind==="reference"?"reference":""}"><td class="method"><span class="rank">${marker}</span><strong>${row.method}</strong><small>${row.version}${links(row)}</small></td><td>${status(row)}</td>${metricCell(row,"fgd",4,ranks)}${metricCell(row,"bc",4,ranks)}${metricCell(row,"diversity",2,ranks)}${metricCell(row,"rotation_geodesic",5,ranks)}${metricCell(row,"expression_l2",5,ranks)}${metricCell(row,"translation_l2",5,ranks)}</tr>`;
+    return `<tr class="${row.kind==="reference"?"reference":""}"><td class="method"><span class="rank">${marker}</span><strong>${row.method}</strong><small>${row.version}${links(row)}</small></td><td>${status(row)}</td>${metricCell(row,"fgd",4,ranks)}${metricCell(row,"bc",4,ranks)}${metricCell(row,"diversity",2,ranks)}${metricCell(row,"utmr_fid",4,ranks)}${metricCell(row,"utmr_paired_distance",4,ranks)}${metricCell(row,"rotation_geodesic",5,ranks)}${metricCell(row,"expression_l2",5,ranks)}${metricCell(row,"translation_l2",5,ranks)}</tr>`;
   }).join("");
   renderPaperParity(data);
 }

speech_to_gesture_results.json

@@ -47,7 +47,10 @@
         "FGD": -0.00048000000001593236,
         "BC": 0.6927973445144807,
         "Diversity": 13.118996183234406
-      }
+      },
+      "utmr_fid": 0.0,
+      "utmr_paired_distance": 0.0,
+      "utmr_diversity": 0.479398316014079
     },
     {
       "method": "Language of Motion",
@@ -93,6 +96,22 @@
           "padding": "zero_motion_code_to_test_length",
           "output_trim": "paired_reference_frame_count"
         }
+      },
+      "utmr_fid": 0.0626219123923138,
+      "utmr_paired_distance": 0.45755861719951857,
+      "utmr_diversity": 0.40607239198027284,
+      "utmr_gt_diversity": 0.479398316014079,
+      "utmr_protocol": {
+        "num_windows": 56,
+        "input_representation": "canonical_smpl22_joints66",
+        "embedding_normalization": "per_sample_l2",
+        "fps": 30.0,
+        "max_window_seconds": 20.0,
+        "min_window_seconds": 0.5,
+        "uses_text": false,
+        "uses_expression": false,
+        "uses_raw_smplx_translation": false,
+        "root_trajectory": "retained_in_fk_joints_after_first_frame_origin"
       }
     },
     {
@@ -105,16 +124,16 @@
       "checkpoint": "https://huggingface.co/ZeyuLing/Motius-EMAGE-BEAT2",
       "model_card": "https://github.com/ZeyuLing/Motius/blob/main/docs/model_zoo/emage.md",
       "paper": "https://arxiv.org/abs/2401.00374",
-      "fgd": 6.19906299336642,
+      "fgd": 6.198973656352552,
       "bc": 7.564133316712631,
-      "diversity": 12.475604227512221,
+      "diversity": 12.475604236570534,
       "rotation_geodesic": 0.27046809122906024,
       "expression_l2": 4.743729881237994,
       "translation_l2": 3.2984432561061547,
       "raw_metrics": {
-        "FGD": 0.619906299336642,
+        "FGD": 0.6198973656352552,
         "BC": 0.7564133316712631,
-        "Diversity": 12.475604227512221
+        "Diversity": 12.475604236570534
       },
       "generation_audit": {
         "max_static_tail_frames": 1,
@@ -162,6 +181,84 @@
           "motion_fps": 30,
           "output_trim": "paired_reference_frame_count"
         }
+      },
+      "utmr_fid": 0.05229689071202964,
+      "utmr_paired_distance": 0.38855470181008667,
+      "utmr_diversity": 0.34108143528053875,
+      "utmr_gt_diversity": 0.479398316014079,
+      "utmr_protocol": {
+        "num_windows": 56,
+        "input_representation": "canonical_smpl22_joints66",
+        "embedding_normalization": "per_sample_l2",
+        "fps": 30.0,
+        "max_window_seconds": 20.0,
+        "min_window_seconds": 0.5,
+        "uses_text": false,
+        "uses_expression": false,
+        "uses_raw_smplx_translation": false,
+        "root_trajectory": "retained_in_fk_joints_after_first_frame_origin"
+      }
+    },
+    {
+      "method": "DiffuseStyleGesture+",
+      "version": "official DSG+ BEAT v0 checkpoint",
+      "kind": "method",
+      "verified": true,
+      "submission_status": "official_checkpoint_verified",
+      "paper_table_eligible": false,
+      "checkpoint": "https://huggingface.co/ZeyuLing/Motius-DiffuseStyleGesture-BEAT",
+      "model_card": "https://github.com/ZeyuLing/Motius/blob/main/docs/model_zoo/diffusestylegesture.md",
+      "paper": "https://arxiv.org/abs/2308.13879",
+      "fgd": 93.7474093069552,
+      "bc": 6.332534812705944,
+      "diversity": 12.020876579601168,
+      "rotation_geodesic": 0.3659869673414887,
+      "expression_l2": 10.92270681552773,
+      "translation_l2": 1.4772545935086279,
+      "raw_metrics": {
+        "FGD": 9.37474093069552,
+        "BC": 0.6332534812705944,
+        "Diversity": 12.020876579601168
+      },
+      "generation_audit": {
+        "max_static_tail_frames": 1,
+        "minimum_active_coverage": 0.9993169398907104,
+        "checkpoint_provenance": {
+          "model001080000.pt": {
+            "bytes": 62077405,
+            "sha256": "2e25d8b257b44e5110a61713476a146e99c34926c763b5dd1c18a6884a857c77"
+          }
+        },
+        "inference_profile": {
+          "protocol": "official_DiffuseStyleGesturePlus_BEAT_v0",
+          "official_profile": true,
+          "runtime_revision": "9649995cdd0b8ce0d42bd02567b2bd03829a0540",
+          "window_frames": 150,
+          "seed_frames": 30,
+          "stride_frames": 120,
+          "diffusion_skip_timesteps": 0,
+          "speaker_id": 2,
+          "base_seed": 123456,
+          "motion_fps": 30,
+          "output_conversion": "official_30fps_BVH_to_FreeTalker_SMPLX_stride1",
+          "output_trim": "paired_reference_frame_count"
+        }
+      },
+      "utmr_fid": 1.6601503728979607,
+      "utmr_paired_distance": 1.2995585722982474,
+      "utmr_diversity": 0.25677656092044593,
+      "utmr_gt_diversity": 0.479398316014079,
+      "utmr_protocol": {
+        "num_windows": 56,
+        "input_representation": "canonical_smpl22_joints66",
+        "embedding_normalization": "per_sample_l2",
+        "fps": 30.0,
+        "max_window_seconds": 20.0,
+        "min_window_seconds": 0.5,
+        "uses_text": false,
+        "uses_expression": false,
+        "uses_raw_smplx_translation": false,
+        "root_trajectory": "retained_in_fk_joints_after_first_frame_origin"
       }
     }
   ]