3 views

Friedrich-M/UniMate

[SIGGRAPH Asia 2026] UniMate: One Unified Model to Animate Diverse Skeletons

Python★ 1,123+217 that day#10 on trending

Analyzed October 2, 2026 at 2c5b384 (main) with jscpd 5.4.0 in 0.1s.

C69.3/100
Project health37,993 lines of code (M)
Duplication72
4.1% in python, bash, perl
Dead code98
0.1%
Complexity48
54%

The score is jscpd's own --health model: markup, data and text left out, tests kept. The numbers below are the trending scan's.

4.06%duplicated code · 1,540 of 37,949 lines
113clones in code
142code files of 159 scanned
5.24%duplicated over all files · 132 clones

The code number leaves out tests, fixtures, docs, examples, data files, styles, translations and vendored code.

Largest clones in the code

  • 33 lines in 2 placespython359 tokens
    238new_motion = np.zeros((n_frames, 1, D), dtype=motion.dtype)
    239new_motion[:, 0, 3:9] = motion[:, ji, 3:9]
    240
    241aug['motion'] = np.concatenate(
    242    [motion[:, :ji], new_motion, motion[:, ji:]], axis=1,
    243)
    244aug['offsets'] = np.concatenate(
    245    [offsets[:ji], offset_new[None], offsets[ji:]], axis=0,
    246)
    247aug['offsets'][ji + 1] = offset_ji_rem
    

    First 10 of 33 lines.

  • 40 lines in 2 placespython295 tokens
    102                         "mesh_animation/animate_mixamo.py.")
    103parser.add_argument("--output_dir", type=str, required=True,
    104                    help="Output directory.")
    105parser.add_argument("--scale", type=float, default=1.0,
    106                    help="Scene normalization scale.")
    107parser.add_argument("--camera_dist", type=float, default=1.5,
    108                    help="Camera distance from origin.")
    109parser.add_argument("--resolution", type=int, default=512,
    110                    help="Render resolution (square).")
    111parser.add_argument("--fps", type=int, default=RENDER_FPS,
    

    First 10 of 40 lines.

  • 33 lines in 2 placespython248 tokens
    147(shift_s, scale_s, gate_s,
    148 shift_t, scale_t, gate_t,
    149 shift_m, scale_m, gate_m) = self.adaLN_modulation(c).chunk(9, dim=-1)
    150
    151def expand_to_x(z):
    152    return z[:, None, None, :]
    153
    154use_ckpt = self.gradient_checkpointing and self.training
    155
    156# spatial attn
    

    First 10 of 33 lines.

7 more
Largest clones over all files, tests and data included
By format, code only
FormatFilesLinesClonesDuplicated linesDuplication
python10235,2861081,5004.25%
bash392,4895401.61%
perl1174000%
By format, all files
FormatFilesLinesClonesDuplicated linesDuplication
python10235,2861081,5004.25%
json97201955076.39%
bash412,5875401.55%
markdown2968000%
perl1174000%
txt189000%
mermaid131000%
text224000%

Trending appearances

DayRankStarsFilesLinesClonesAll filesCodeHealthCommit
Oct 2, 2026#101,123 (+217)15939,8791325.24%4.06%C 69.32c5b384

Want this for your own project? Install jscpd and run jscpd ., or measure it the way the baselines were measured.