SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 1908.07201 · 2019

Three-D Safari: Learning to Estimate Zebra Pose, Shape, and Texture from Images "In the Wild"

arXiv · PDF · Open in the Atlas

Code that ran

12 of 17 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 12 have no contract verdict; 0 failures of our instrument; 5 no recorded run), measured 2026-11-21; 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/1908.07201.json · badge: /api/ran/1908.07201.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
silviazuffi/smalst canonical (classifier label)
classifier basis: census
12 ran · 5 no recorded run
FunctionStatusWhere it lives
batch_rodrigues Ran
ran on a synthesized input
silviazuffi/smalst/smal_model/batch_lbs.py
code served (permissive licence) · get_code("eb25cab46005f07e")
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 3, dtype family float, shape (2, 3, 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
import numpy as np

torch.manual_seed(0)

class OptsStandIn:
    def __init__(self):
        self.gpu_id = 0

# Create a small batch of 3D rotation vectors (Nx3)
batch_size = 2
theta = torch.randn(batch_size, 3)

# Create options object with gpu_id attribute
opts = OptsStandIn()

ARGS = {
    "theta": theta,
    "opts": opts
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Tensor, ndim 3, dtype family float, shape (2, 3, 3) — shape (2, 3, 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
batch_skew Ran
ran on a synthesized input
silviazuffi/smalst/smal_model/batch_lbs.py
code served (permissive licence) · get_code("d7418e81eccf569d")
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 3, dtype family float, shape (2, 3, 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
import numpy as np

torch.manual_seed(0)

class OptsStandIn:
    def __init__(self):
        self.gpu_id = 0

vec = torch.randn(2, 3)
batch_size = 2
opts = OptsStandIn()

ARGS = {
    "vec": vec,
    "batch_size": batch_size,
    "opts": opts,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Tensor, ndim 3, dtype family float, shape (2, 3, 3) — shape (2, 3, 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
convert_as Ran
ran on a synthesized input
silviazuffi/smalst/nnutils/nmr.py
code served (permissive licence) · get_code("7667d599b9606913")
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. 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
import torch.nn as nn

torch.manual_seed(0)

# Create source tensor - small CPU tensor with dtype float32
src = torch.randn(2, 4, dtype=torch.float32)

# Create target tensor - small CPU tensor with different dtype
trg = torch.randn(3, 4, dtype=torch.float64)

ARGS = {
    "src": src,
    "trg": trg,
}
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
fc Ran
ran on a synthesized input
silviazuffi/smalst/nnutils/net_blocks.py
code served (permissive licence) · get_code("483c5935ef25b6f6")
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 Sequential. 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
import torch.nn as nn

torch.manual_seed(0)

ARGS = {
    'norm_type': 'batch',
    'nc_inp': 4,
    'nc_out': 3
}
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 Sequential — a Sequential output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a Sequential output the recorded ndim 1 and dtype family float describe that coercion, not the output, and are not 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
fc_stack Ran
ran on a synthesized input
silviazuffi/smalst/nnutils/net_blocks.py
code served (permissive licence) · get_code("e59492399c2004d7")
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 Sequential. 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
import torch.nn as nn

torch.manual_seed(0)

ARGS = {
    'nc_inp': 16,
    'nc_out': 32,
    'nlayers': 3,
    'norm_type': 'batch'
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Sequential — a Sequential output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a Sequential output the recorded ndim 2 and dtype family float describe that coercion, not the output, and are not 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
fc_stack_dropout Ran
ran on a synthesized input
silviazuffi/smalst/nnutils/net_blocks.py
code served (permissive licence) · get_code("989712660a666ac2")
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 Sequential. 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
import torch.nn as nn

torch.manual_seed(0)

ARGS = {
    "nc_inp": 16,
    "nc_out": 8,
    "nlayers": 3,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Sequential — a Sequential output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a Sequential output the recorded ndim 1 and dtype family float describe that coercion, not the output, and are not 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
mirror_keypoints Ran
ran on a synthesized input
silviazuffi/smalst/smal_eval.py
code served (permissive licence) · get_code("02bff3f77553f7b4")
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 numpy as np

# Create small valid keypoint arrays
# The function expects keyp to be a 2D array with shape (28, 2) based on the dx2sx mapping
# and the indexing pattern keyp[dx2sx, 0] and keyp[dx2sx, 1]
keyp = np.random.rand(28, 2) * 100  # 28 keypoints with (x, y) coordinates
vis = np.random.randint(0, 2, size=28)  # visibility flags for each keypoint

img_w = 640  # image width for mirror transformation

ARGS = {
    "keyp": keyp,
    "vis": vis,
    "img_w": img_w
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy — 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
perspective_proj_withz Ran
ran on a synthesized input
silviazuffi/smalst/nnutils/geom_utils.py
code served (permissive licence) · get_code("42874ab7b0a7db85")
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 3, dtype family float, shape (2, 4, 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)

# X: B x N x 3 (batch_size=2, num_points=4, coords=3)
X = torch.randn(2, 4, 3)

# cam: B x 3: [f, cx, cy]
cam = torch.randn(2, 3)

# offset_z: scalar, not used according to docstring
offset_z = 0

# cuda_device: int
cuda_device = 0

# norm_f: scalar multiplier for focal length
norm_f = 1.0

# norm_z: scalar offset for z coordinate
norm_z = 0.0

# norm_f0: scalar base focal length
norm_f0 = 0.0

ARGS = {
    "X": X,
    "cam": cam,
    "offset_z": offset_z,
    "cuda_device": cuda_device,
    "norm_f": norm_f,
    "norm_z": norm_z,
    "norm_f0": norm_f0,
}
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 3, dtype family float, shape (2, 4, 3) — shape (2, 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
sample_textures Ran
ran on a synthesized input
silviazuffi/smalst/nnutils/geom_utils.py
code served (permissive licence) · get_code("cafb397ac53b9cd9")
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 5, dtype family float, shape (2, 3, 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
import torch.nn.functional as F

torch.manual_seed(0)

# texture_flow: B x F x T x T x 2
# B = batch size = 2
# F = number of frames = 3
# T = texture size = 4
# Last dimension = 2 (normalized coordinates)
B, F, T = 2, 3, 4
texture_flow = torch.randn(B, F, T, T, 2) * 0.5  # Keep in reasonable range for grid_sample

# images: B x 3 x N x N
# B = batch size = 2
# 3 = RGB channels
# N = image size = 8
N = 8
images = torch.randn(B, 3, N, N)

ARGS = {
    "texture_flow": texture_flow,
    "images": images,
}
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 5, dtype family float, shape (2, 3, 4, 4, 3) — shape (2, 3, 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
texture_loss Ran
ran on a synthesized input
silviazuffi/smalst/nnutils/loss_utils.py
code served (permissive licence) · get_code("3036961b6f0ad211")
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 small tensors with distinct dimensions
# B=2 (batch), C=3 (channels), H=4 (height), W=5 (width)
B, C, H, W = 2, 3, 4, 5

img_pred = torch.randn(B, C, H, W)
img_gt = torch.randn(B, C, H, W)

# Masks: B x H x W
mask_pred = torch.randint(0, 2, (B, H, W)).float()
mask_gt = torch.randint(0, 2, (B, H, W)).float()

ARGS = {
    "img_pred": img_pred,
    "img_gt": img_gt,
    "mask_pred": mask_pred,
    "mask_gt": mask_gt,
}
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
undo_chumpy Ran
ran on a synthesized input
silviazuffi/smalst/smal_model/smal_torch.py
code served (permissive licence) · get_code("e24744b154d5aa09")
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 (2, 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

class ChumbyStandIn:
    """Mock object that has a .r attribute for the non-ndarray case"""
    def __init__(self, data):
        self.r = data

# Test case 1: numpy array input
x_numpy = np.array([[1, 2, 3], [4, 5, 6]])

# Test case 2: object with .r attribute
x_chumpy = ChumbyStandIn(np.array([[7, 8, 9], [10, 11, 12]]))

# Create ARGS dict - we'll use the numpy array case as the primary test
ARGS = {
    "x": x_numpy
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
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 (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
uv_flow_loss Ran
ran on a synthesized input
silviazuffi/smalst/nnutils/loss_utils.py
code served (permissive licence) · get_code("44802711cbcd9f9a")
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 small tensors with distinct dimensions for clarity
B = 2  # batch size
H = 4  # height
W = 3  # width

# uv_flow_pred: B x 2 x H x W
uv_flow_pred = torch.randn(B, 2, H, W)

# uv_flow_gt_w_mask: B x 3 x H x W
# First 2 channels are ground truth UV flow, 3rd channel is mask
uv_flow_gt_w_mask = torch.randn(B, 3, H, W)

ARGS = {
    "uv_flow_pred": uv_flow_pred,
    "uv_flow_gt_w_mask": uv_flow_gt_w_mask,
}
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
align_smal_template_to_symmetry_axis Not yet run
unverified (harvested, no recorded run)
silviazuffi/smalst/smal_model/smal_basics.py
code served (permissive licence) · get_code("fffc41d90e743fa6")
How it ran

no invocation recorded.

batch_lrotmin Not yet run
unverified (harvested, no recorded run)
silviazuffi/smalst/smal_model/batch_lbs.py
code served (permissive licence) · get_code("4f447281537c79d4")
How it ran

no invocation recorded.

get_bbox Not yet run
unverified (harvested, no recorded run)
silviazuffi/smalst/smal_eval.py
code served (permissive licence) · get_code("251a4caf1c684e45")
How it ran

no invocation recorded.

preprocess_image Not yet run
unverified (harvested, no recorded run)
silviazuffi/smalst/smal_eval.py
code served (permissive licence) · get_code("8c9b723cddb2255d")
How it ran

no invocation recorded.

texture_dt_loss Not yet run
unverified (harvested, no recorded run)
silviazuffi/smalst/nnutils/loss_utils.py
code served (permissive licence) · get_code("742394a7d024dbb9")
How it ran

no invocation recorded.

Also present in other papers' code

Of 17 samples lifted from this paper's linked repositories, 17 carry a function identity and 0 do not. Of the 17 distinct functions with an identity, 5 are also present in other papers whose code we hold (0 with at least one paper dated earlier, 1 the same month, 5 later) and 12 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
fc 2 other papers
2 later
2007.11110 (2020-07)
Who Left the Dogs Out? 3D Animal Reconstruction with Expectation Maximization in the Loop
2111.08176 (2021-11)
Coarse-to-fine Animal Pose and Shape Estimation
fc_stack 2 other papers
2 later
2007.11110 (2020-07)
Who Left the Dogs Out? 3D Animal Reconstruction with Expectation Maximization in the Loop
2111.08176 (2021-11)
Coarse-to-fine Animal Pose and Shape Estimation
fc_stack_dropout 2 other papers
2 later
2007.11110 (2020-07)
Who Left the Dogs Out? 3D Animal Reconstruction with Expectation Maximization in the Loop
2111.08176 (2021-11)
Coarse-to-fine Animal Pose and Shape Estimation
undo_chumpy 2 other papers
1 the same month, 1 later
1908.04781 (2019-08)
Predicting 3D Human Dynamics from Video
2007.11110 (2020-07)
Who Left the Dogs Out? 3D Animal Reconstruction with Expectation Maximization in the Loop
batch_lrotmin 1 other paper
1 later
2007.11110 (2020-07)
Who Left the Dogs Out? 3D Animal Reconstruction with Expectation Maximization in the Loop

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 2026-11-21.

Repositories linked to this paper

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

Abstract

We present the first method to perform automatic 3D pose, shape and texture capture of animals from images acquired in-the-wild. In particular, we focus on the problem of capturing 3D information about Grevy's zebras from a collection of images. The Grevy's zebra is one of the most endangered species in Africa, with only a few thousand individuals left. Capturing the shape and pose of these animals can provide biologists and conservationists with information about animal health and behavior. In contrast to research on human pose, shape and texture estimation, training data for endangered species is limited, the animals are in complex natural scenes with occlusion, they are naturally camouflaged, travel in herds, and look similar to each other. To overcome these challenges, we integrate the recent SMAL animal model into a network-based regression pipeline, which we train end-to-end on synthetically generated images with pose, shape, and background variation. Going beyond state-of-the-art methods for human shape and pose estimation, our method learns a shape space for zebras during training. Learning such a shape space from images using only a photometric loss is novel, and the approach can be used to learn shape in other settings with limited 3D supervision. Moreover, we couple 3D pose and shape prediction with the task of texture synthesis, obtaining a full texture map of the animal from a single image. We show that the predicted texture map allows a novel per-instance unsupervised optimization over the network features. This method, SMALST (SMAL with learned Shape and Texture) goes beyond previous work, which assumed manual keypoints and/or segmentation, to regress directly from pixels to 3D animal shape, pose and texture. Code and data are available at https://github.com/silviazuffi/smalst.

For agents

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

get_harvested_code_for_paper("1908.07201")
get_code_for_paper("1908.07201")
have("1908.07201")

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

curl https://syntology.ai/api/ran/1908.07201.json

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

[![Syntology run record](https://syntology.ai/api/ran/1908.07201.svg)](https://syntology.ai/paper/1908.07201)

Connect an agent — have() is free.