SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2308.08545 · 2023

TeCH: Text-guided Reconstruction of Lifelike Clothed Humans

arXiv · PDF · Open in the Atlas

Code that ran

12 of 14 samples (functions and classes) lifted from this paper's linked repositories executed on a synthesized input in Syntology's jail (0 of them constructed an object rather than computing a result; 0 honoured the contract we drafted, 0 violated it and 0 have no contract verdict; 12 failures of our instrument; 2 no recorded run), measured 2026-11-22; repository links from the Papers with Code archive attributed under CC BY-SA, no affiliation; bodies inline only under a permissive licence as GitHub detects it. Not a reproduction of the paper's results.

"Ran" means the function executed on a synthesized input and returned a value. It is not a reproduction of the paper's results. This record, dated, as JSON: /api/ran/2308.08545.json · badge: /api/ran/2308.08545.svg

Each repository's role is a label: a classifier's output recorded on the link, or, where no role is recorded, what the Papers with Code archive marked. The line under it names its basis. Neither says who wrote the code.

RepositoryRoleRan
huangyangyi/tech canonical (classifier label)
classifier basis: census
12 ran · 2 no recorded run
FunctionStatusWhere it lives
compact_tets Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
huangyangyi/tech/core/lib/dmtet_network.py
code served (permissive licence) · get_code("21f8db18059249bd")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned tuple. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch

torch.manual_seed(0)

# Create small tensors to satisfy the function requirements
# pos_nx3: n vertices, each with 3 coordinates
n_vertices = 6
pos_nx3 = torch.randn(n_vertices, 3)

# sdf_n: signed distance field values for each vertex
sdf_n = torch.randn(n_vertices)

# tet_fx4: f tetrahedra, each defined by 4 vertex indices
n_tets = 4
tet_fx4 = torch.tensor([
    [0, 1, 2, 3],
    [1, 2, 3, 4],
    [2, 3, 4, 5],
    [0, 1, 4, 5]
], dtype=torch.long)

ARGS = {
    'pos_nx3': pos_nx3,
    'sdf_n': sdf_n,
    'tet_fx4': tet_fx4
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type tuple — a tuple output has no array shape of its own; none is stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
crop_by_mask Ran
ran; our contract draft was wrong, not the code
huangyangyi/tech/core/lib/loss_utils.py
code served (permissive licence) · get_code("720478a23c7f1394")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned tuple. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch

torch.manual_seed(0)

# Create small tensors for rgb and alpha
# rgb shape: (batch, channels, height, width) = (2, 3, 64, 64)
# alpha shape: (batch, channels, height, width) = (2, 1, 64, 64)
# The function accesses alpha[0, 0] so we need at least 2 dimensions after indexing
rgb = torch.rand(2, 3, 64, 64)
alpha = torch.rand(2, 1, 64, 64)

# Make sure alpha has some non-zero values so the mask isn't empty
alpha[0, 0, 20:50, 20:50] = 0.5

ARGS = {
    "rgb": rgb,
    "alpha": alpha,
    "base_size": 64
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type tuple — a tuple output has no array shape of its own; none is stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
get_edt Ran
ran; our contract draft was wrong, not the code
huangyangyi/tech/core/lib/loss_utils.py
code served (permissive licence) · get_code("f74a827acce57b00")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned Tensor, ndim 4, dtype family float, shape (2, 1, 4, 3) as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
import torch.nn.functional as F
from scipy.ndimage import distance_transform_edt

torch.manual_seed(0)

# Create a small binary mask tensor (batch_size=2, channels=1, height=4, width=3)
gt_mask = torch.randint(0, 2, (2, 1, 4, 3), dtype=torch.float32)

# Create a loss mask (optional, same shape as gt_mask)
loss_mask = torch.randint(0, 2, (2, 1, 4, 3), dtype=torch.float32)

ARGS = {
    "gt_mask": gt_mask,
    "loss_mask": loss_mask,
    "kernel_size": 7,
    "edt_power": 0.25,
    "l2_weight": 0.01,
    "edge_weight": 0.01,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
scipy, torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Tensor, ndim 4, dtype family float, shape (2, 1, 4, 3) — shape (2, 1, 4, 3) as a later run recorded it
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
get_embedder Ran
ran; our contract draft was wrong, not the code
huangyangyi/tech/core/lib/network_utils.py
code served (permissive licence) · get_code("d5b158881d454655")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1) and returned tuple, ndim 1, dtype family float; no output shape is stated. The image it ran in was not recorded. If your copy returns the same type and ndim on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch

torch.manual_seed(0)

ARGS = {
    'multires': 4
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type tuple, ndim 1, dtype family float — a tuple output has no array shape of its own; none is stated
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
get_encoder Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
huangyangyi/tech/core/lib/encoding.py
code served (permissive licence) · get_code("f8e889919d6ffe62")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned tuple, ndim 1, dtype family float; no output shape is stated. The image it ran in was not recorded. If your copy returns the same type and ndim on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
import torch.nn as nn

torch.manual_seed(0)

ARGS = {
    "encoding": "frequency_torch",
    "input_dim": 3,
    "multires": 6,
    "degree": 4,
    "num_levels": 16,
    "level_dim": 2,
    "base_resolution": 16,
    "log2_hashmap_size": 19,
    "desired_resolution": 2048,
    "align_corners": False,
    "interpolation": "linear",
    "input_bounds": None,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
freqencoder, gridencoder, shencoder, torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type tuple, ndim 1, dtype family float — a tuple output has no array shape of its own; none is stated
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
get_encoder Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
huangyangyi/tech/core/lib/network_utils.py
code served (permissive licence) · get_code("cd89be610e2d21cd")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned tuple, ndim 1, dtype family float; no output shape is stated. The image it ran in was not recorded. If your copy returns the same type and ndim on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
import torch.nn as nn

torch.manual_seed(0)

ARGS = {
    'encoding': 'frequency_torch',
    'input_dim': 3,
    'multires': 6,
    'degree': 4,
    'num_levels': 16,
    'level_dim': 2,
    'base_resolution': 16,
    'log2_hashmap_size': 19,
    'desired_resolution': 2048,
    'align_corners': False,
    'interpolation': 'linear',
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
shencoder, torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type tuple, ndim 1, dtype family float — a tuple output has no array shape of its own; none is stated
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
laplace_regularizer_const Ran
ran; our contract draft was wrong, not the code
huangyangyi/tech/core/lib/dmtet_network.py
code served (permissive licence) · get_code("39c9419cae242513")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned Tensor, ndim 0, dtype family float, shape () as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch

torch.manual_seed(0)

# Create a simple mesh with 4 vertices and 2 triangular faces
# mesh_verts: (num_vertices, 3) - 3D coordinates
# mesh_faces: (num_faces, 3) - vertex indices for each face

mesh_verts = torch.randn(4, 3, dtype=torch.float32)
mesh_faces = torch.tensor([
    [0, 1, 2],
    [1, 2, 3]
], dtype=torch.long)

ARGS = {
    "mesh_verts": mesh_verts,
    "mesh_faces": mesh_faces,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Tensor, ndim 0, dtype family float, shape () — shape () as a later run recorded it
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
nms Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
huangyangyi/tech/core/lib/annotators.py
code served (permissive licence) · get_code("fa18e29c0ab322c6")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned ndarray, ndim 2, dtype family int, shape (4, 3) as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import numpy as np
import cv2

# Create a small test image (4x3 to have distinct dimensions)
x = np.random.randint(0, 256, size=(4, 3), dtype=np.uint8)

# Threshold value - any reasonable float
t = 50.0

# Sigma for Gaussian blur - controls the blur amount
s = 1.5

ARGS = {
    "x": x,
    "t": t,
    "s": s
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
cv2, numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type ndarray, ndim 2, dtype family int, shape (4, 3) — shape (4, 3) as a later run recorded it
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
rgb2srgb Ran
ran; our contract draft was wrong, not the code
huangyangyi/tech/core/lib/color_utils.py
code served (permissive licence) · get_code("29830e1e724bb23b")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned Tensor, ndim 2, dtype family float, shape (2, 4) as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch

torch.manual_seed(0)

# Create a small tensor with values in the typical range for RGB images (0-1)
# Using batch size 2, with 4 color channels as a small test case
imgs = torch.rand(2, 4) * 0.5  # Values between 0 and 0.5 to test both branches of the where condition

ARGS = {
    "imgs": imgs
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Tensor, ndim 2, dtype family float, shape (2, 4) — shape (2, 4) as a later run recorded it
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
rgb2xyz Ran
ran; our contract draft was wrong, not the code
huangyangyi/tech/core/lib/color_utils.py
code served (permissive licence) · get_code("3c6c1b5fa43f8aa8")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned Tensor, ndim 2, dtype family float, shape (2, 3) as a later run recorded it. The image it ran in was not recorded. The code mentions cuda, and the harness aliases cuda to the CPU for every sample; whether this run needed that is not recorded. If your copy returns the same type, ndim and shape on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch

torch.manual_seed(0)

# The function expects:
# - var: a tensor that will be matrix-multiplied with a 3x3 transform matrix
#   The function does torch.matmul(var, transform.t()), so var should have shape (..., 3)
#   to match the last dimension with transform.t() which is (3, 3)
# - device: a string specifying device (default 'cuda')

# Create a small RGB tensor with shape (2, 3) - batch of 2 RGB values
rgb_tensor = torch.randn(2, 3)

# device parameter - use 'cpu' instead of 'cuda' to avoid CUDA errors
device_str = 'cpu'

ARGS = {
    'var': rgb_tensor,
    'device': device_str
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Tensor, ndim 2, dtype family float, shape (2, 3) — shape (2, 3) as a later run recorded it
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
rgb2ycrcb Ran
ran; our contract draft was wrong, not the code
huangyangyi/tech/core/lib/color_utils.py
code served (permissive licence) · get_code("99a27d0a0d932250")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned Tensor, ndim 4, dtype family float, shape (2, 4, 4, 3) as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch

torch.manual_seed(0)

# The function expects imgs to be a tensor with shape (..., 3)
# where the last dimension contains RGB channels in range [0, 1]
# Let's create a small batch of 2 images with 4x4 spatial dimensions
imgs = torch.rand(2, 4, 4, 3)

ARGS = {
    "imgs": imgs
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Tensor, ndim 4, dtype family float, shape (2, 4, 4, 3) — shape (2, 4, 4, 3) as a later run recorded it
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
safe_normalize Ran
ran; our contract draft was wrong, not the code
huangyangyi/tech/core/lib/camera_utils.py
code served (permissive licence) · get_code("1c9cf033d080b2e0")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned Tensor, ndim 2, dtype family float, shape (2, 4) as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch

torch.manual_seed(0)

# The function safe_normalize takes a tensor x and normalizes it
# by dividing by the L2 norm computed along the last dimension
# Parameters:
# - x: a tensor of any shape
# - eps: a small epsilon value for numerical stability (default 1e-20)

# Create a small test tensor with shape (2, 4)
# This gives us a batch of 2 vectors, each of dimension 4
x = torch.randn(2, 4)

# eps uses the default value from the function signature
eps = 1e-20

ARGS = {
    "x": x,
    "eps": eps,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Tensor, ndim 2, dtype family float, shape (2, 4) — shape (2, 4) as a later run recorded it
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
get_rays Not yet run
unverified (harvested, no recorded run)
huangyangyi/tech/core/lib/camera_utils.py
code served (permissive licence) · get_code("33fea52680be893b")
How it ran

no invocation recorded.

silhouette_loss Not yet run
unverified (harvested, no recorded run)
huangyangyi/tech/core/lib/loss_utils.py
code served (permissive licence) · get_code("2396978f462745b3")
How it ran

no invocation recorded.

Also present in other papers' code

Of 14 samples lifted from this paper's linked repositories, 14 carry a function identity and 0 do not. Of the 14 distinct functions with an identity, 3 are also present in other papers whose code we hold (2 with at least one paper dated earlier, 0 the same month, 2 later) and 11 are only here among the 52,977 papers whose code Syntology holds (as of 2026-09-26).

Only here means: among the 52,977 papers whose code Syntology holds (51,887 with lifted samples, 1,090 through their assemblies only), as of 2026-09-26, no other paper's lifted samples or current assembly bricks contain an identical function body after normalisation.

FunctionAlso present inEarliest-datedOthers named
nms 6 other papers
6 later
2312.04524 (2023-12)
RAVE: Randomized Noise Shuffling for Fast and Consistent Video Editing with Diffusion Models
2312.07315 (2023-12)
NVS-Adapter: Plug-and-Play Novel View Synthesis from a Single Image
2312.11392 (2023-12)
SCEdit: Efficient and Controllable Image Diffusion Generation via Skip Connection Editing
and 3 more not named here
safe_normalize 6 other papers
2 dated earlier, 4 later
2209.14988 (2022-09)
DreamFusion: Text-to-3D using 2D Diffusion
2211.07600 (2022-11)
Latent-NeRF for Shape-Guided Generation of 3D Shapes and Textures
2312.00778 (2023-12)
MorpheuS: Neural Dynamic 360° Surface Reconstruction from Monocular RGB-D Video
and 3 more not named here
get_embedder 1 other paper
1 dated earlier
1911.05063 (2019-11)
Kaolin: A PyTorch Library for Accelerating 3D Deep Learning Research

Presence, not direction: an identical function body in two papers' repositories does not say who wrote it, which came from which, or that either paper's method rests on it. Matching is exact after normalisation, so a renamed or edited function is not matched. Dates are when papers appeared, not when code was written. Only .py files the harvester lifted are covered; notebooks are not read. A new-style arXiv id YYMM.NNNN dates a paper to the month 20YY-MM; otherwise Paper.year dates it to the year; otherwise it is undated. Dates are when the paper appeared, not when the code was written. Read 2026-11-22.

Repositories linked to this paper

Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.

Abstract

Despite recent research advancements in reconstructing clothed humans from a single image, accurately restoring the "unseen regions" with high-level details remains an unsolved challenge that lacks attention. Existing methods often generate overly smooth back-side surfaces with a blurry texture. But how to effectively capture all visual attributes of an individual from a single image, which are sufficient to reconstruct unseen areas (e.g., the back view)? Motivated by the power of foundation models, TeCH reconstructs the 3D human by leveraging 1) descriptive text prompts (e.g., garments, colors, hairstyles) which are automatically generated via a garment parsing model and Visual Question Answering (VQA), 2) a personalized fine-tuned Text-to-Image diffusion model (T2I) which learns the "indescribable" appearance. To represent high-resolution 3D clothed humans at an affordable cost, we propose a hybrid 3D representation based on DMTet, which consists of an explicit body shape grid and an implicit distance field. Guided by the descriptive prompts + personalized T2I diffusion model, the geometry and texture of the 3D humans are optimized through multi-view Score Distillation Sampling (SDS) and reconstruction losses based on the original observation. TeCH produces high-fidelity 3D clothed humans with consistent & delicate texture, and detailed full-body geometry. Quantitative and qualitative experiments demonstrate that TeCH outperforms the state-of-the-art methods in terms of reconstruction accuracy and rendering quality. The code will be publicly available for research purposes at https://huangyangyi.github.io/TeCH

For agents

The same record, over MCP at https://syntology.ai/mcp:

get_harvested_code_for_paper("2308.08545")
get_code_for_paper("2308.08545")
have("2308.08545")

The run record, dated, one paper per request, free:

curl https://syntology.ai/api/ran/2308.08545.json

A badge for a README (the split and the date, never a ratio):

[![Syntology run record](https://syntology.ai/api/ran/2308.08545.svg)](https://syntology.ai/paper/2308.08545)

Connect an agent — have() is free.