SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2104.02273 · CVPR · 2021

Multi-View Multi-Person 3D Pose Estimation with Plane Sweep Stereo

Hee Lee, Jiahao Gim

arXiv · PDF · Open in the Atlas

Code that ran

6 of 9 samples (functions and classes) lifted from this paper's linked repositories executed on a synthesized input in Syntology's jail (2 of them constructed an object rather than computing a result; 0 honoured the contract we drafted, 0 violated it and 3 have no contract verdict; 3 failures of our instrument; 3 no recorded run), measured 2026-11-22; 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/2104.02273.json · badge: /api/ran/2104.02273.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
jiahaoLjh/PlaneSweepPose marked official in Papers with Code
Papers with Code archive
6 ran · 3 no recorded run
FunctionStatusWhere it lives
JointCNN Ran
ran on a synthesized input · constructed an object; no result was computed
jiahaoLjh/PlaneSweepPose/lib/models/mvmppe.py
code served (permissive licence) · get_code("e441a0ff8dbcb8cc")
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 an instance of its own class, JointCNN: the run constructed the object and did not call it, so no result was computed. 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)

class JointCNN(nn.Module):

    def __init__(self, num_joints, hidden_size, output_size):
        super().__init__()
        self.input_layer = nn.Sequential(nn.Conv1d(num_joints, hidden_size, kernel_size=1, bias=False), nn.BatchNorm1d(hidden_size), nn.ReLU(inplace=True))
        self.res_layers = nn.ModuleList([nn.Sequential(nn.Conv1d(hidden_size, hidden_size, kernel_size=3, padding=1, dilation=1, bias=False), nn.BatchNorm1d(hidden_size), nn.ReLU(inplace=True), nn.Conv1d(hidden_size, hidden_size, kernel_size=3, padding=2, dilation=2, bias=False), nn.BatchNorm1d(hidden_size), nn.ReLU(inplace=True), nn.Conv1d(hidden_size, hidden_size, kernel_size=3, padding=4, dilation=4, bias=False), nn.BatchNorm1d(hidden_size), nn.ReLU(inplace=True), nn.Conv1d(hidden_size, hidden_size, kernel_size=3, padding=8, dilation=8, bias=False), nn.BatchNorm1d(hidden_size), nn.ReLU(inplace=True), nn.Conv1d(hidden_size, hidden_size, kernel_size=3, padding=16, dilation=16, bias=False), nn.BatchNorm1d(hidden_size), nn.ReLU(inplace=True), nn.Conv1d(hidden_size, hidden_size, kernel_size=3, padding=32, dilation=32, bias=False), nn.BatchNorm1d(hidden_size), nn.ReLU(inplace=True), nn.Conv1d(hidden_size, hidden_size, kernel_size=3, padding=1, dilation=1, bias=False), nn.BatchNorm1d(hidden_size), nn.ReLU(inplace=True)) for _ in range(2)])
        self.output_layer = nn.Conv1d(hidden_size, output_size, kernel_size=1, bias=False)

    def forward(self, x):
        """
        Args
            x: [B, Nj, Nd]
        Returns
            [B, Cout, Nd]
        """
        x = self.input_layer(x)
        for l in range(len(self.res_layers)):
            y = self.res_layers[l](x)
            x = x + y
        x = self.output_layer(x)
        return x

ARGS = {
    "num_joints": 4,
    "hidden_size": 8,
    "output_size": 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 JointCNN — the run constructed an instance of its own class and did not call it; no shape is stated — the instrument coerces any output to an array; for a JointCNN output the recorded ndim 0 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
PoseCNN Ran
ran on a synthesized input · constructed an object; no result was computed
jiahaoLjh/PlaneSweepPose/lib/models/mvmppe.py
code served (permissive licence) · get_code("a2e26e651381ea85")
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 an instance of its own class, PoseCNN: the run constructed the object and did not call it, so no result was computed. 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)

# Parameters for PoseCNN.__init__
num_joints = 5  # Number of joints (input channels)
hidden_size = 8  # Hidden dimension
output_size = 3  # Output dimension

ARGS = {
    "num_joints": num_joints,
    "hidden_size": hidden_size,
    "output_size": output_size,
}
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 PoseCNN — the run constructed an instance of its own class and did not call it; no shape is stated — the instrument coerces any output to an array; for a PoseCNN output the recorded ndim 0 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
SoftArgMax Ran
ran on a synthesized input
jiahaoLjh/PlaneSweepPose/lib/models/mvmppe.py
code served (permissive licence) · get_code("2541e2a5d1dc76f9")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, class construct-then-call prompt, model not recorded) and returned Tensor, ndim 2, 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

# INIT_ARGS: SoftArgMax.__init__ takes no parameters beyond self
INIT_ARGS = {}

# ARGS: forward method parameters
# x: [B, C, Nd] - batch=2, channels=3, depth=5
# labels: [Nd] - must match the last dimension of x
# kernel_size: optional, defaults to 0 (which means use x.size(-1))

B = 2
C = 3
Nd = 5

x = torch.randn(B, C, Nd)
labels = torch.randn(Nd)
kernel_size = 0

ARGS = {
    'x': x,
    'labels': labels,
    'kernel_size': kernel_size
}
Setup block drafted by
a language model, class construct-then-call prompt, model not recorded
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 — no output shape is stated: observe_signature.py observed this class with its constructor sizes canonicalised where it could, which the node does not record, so its shape is not stated as this setup block's
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: torch and numpy seeded to 0 by the harness immediately before the class was constructed from INIT_ARGS (fixture_check.py)
torch_back_project_point Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
jiahaoLjh/PlaneSweepPose/lib/models/mvmppe.py
code served (permissive licence) · get_code("a29c52185d0b6e36")
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, 3, 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)

# Parameters for torch_back_project_point
batch_size = 2
num_persons = 3
num_joints = 4
num_dimension = 2

# y: BxPxJx2 points in image frame
y = torch.randn(batch_size, num_persons, num_joints, num_dimension)

# depth: B or BxP
depth = torch.randn(batch_size, num_persons)

# R: Bx3x3 Camera rotation matrix
R = torch.eye(3).unsqueeze(0).repeat(batch_size, 1, 1) + torch.randn(batch_size, 3, 3) * 0.01

# T: Bx3x1 Camera translation parameters
T = torch.randn(batch_size, 3, 1)

# f: Bx2x1 Camera focal length
f = torch.ones(batch_size, 2, 1) * 500.0

# c: Bx2x1 Camera center
c = torch.ones(batch_size, 2, 1) * 256.0

# k: Bx3x1 distortion coefficients
k = torch.randn(batch_size, 3, 1) * 0.01

# p: Bx2x1 tangential distortion
p = torch.randn(batch_size, 2, 1) * 0.001

ARGS = {
    "y": y,
    "depth": depth,
    "R": R,
    "T": T,
    "f": f,
    "c": c,
    "k": k,
    "p": p,
}
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, 3, 4, 3) — shape (2, 3, 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
torch_project_point Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
jiahaoLjh/PlaneSweepPose/lib/models/mvmppe.py
code served (permissive licence) · get_code("e9e64cecb7168c61")
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, 3, 4, 2) 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)

# Define dimensions
batch_size = 2
num_persons = 3
num_joints = 4
num_dimension = 3

# x: BxPxJx3 points in world coordinates
x = torch.randn(batch_size, num_persons, num_joints, num_dimension)

# R: Bx3x3 Camera rotation matrix
R = torch.randn(batch_size, 3, 3)

# T: Bx3x1 Camera translation parameters
T = torch.randn(batch_size, 3, 1)

# f: Bx2x1 Camera focal length
f = torch.randn(batch_size, 2, 1)

# c: Bx2x1 Camera center
c = torch.randn(batch_size, 2, 1)

# k: Bx3x1 Distortion coefficients
k = torch.randn(batch_size, 3, 1)

# p: Bx2x1 Tangential distortion coefficients
p = torch.randn(batch_size, 2, 1)

ARGS = {
    'x': x,
    'R': R,
    'T': T,
    'f': f,
    'c': c,
    'k': k,
    'p': p
}
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, 3, 4, 2) — shape (2, 3, 4, 2) 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
torch_unfold_camera_param Ran
ran; our contract draft was wrong, not the code
jiahaoLjh/PlaneSweepPose/lib/models/mvmppe.py
code served (permissive licence) · get_code("69f36ea163abdf52")
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)

# Construct a camera dictionary with all required keys
# The function expects:
# - R: rotation matrix (3x3)
# - T: translation vector (3,)
# - fx, fy: focal lengths (scalars)
# - cx, cy: principal point (scalars)
# - k: distortion coefficients
# - p: distortion coefficients

camera = {
    'R': torch.randn(3, 3),
    'T': torch.randn(3),
    'fx': torch.tensor(500.0),
    'fy': torch.tensor(510.0),
    'cx': torch.tensor(320.0),
    'cy': torch.tensor(240.0),
    'k': torch.randn(4),
    'p': torch.randn(2)
}

ARGS = {
    'camera': camera,
    'device': None
}
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
MultiViewMultiPersonPoseNet Not yet run
unverified (harvested, no recorded run)
jiahaoLjh/PlaneSweepPose/lib/models/mvmppe.py
code served (permissive licence) · get_code("3d720292b6c7f203")
How it ran

no invocation recorded.

torch_back_project_pose Not yet run
unverified (harvested, no recorded run)
jiahaoLjh/PlaneSweepPose/lib/models/mvmppe.py
code served (permissive licence) · get_code("fd94ab017d8d8274")
How it ran

no invocation recorded.

torch_project_pose Not yet run
unverified (harvested, no recorded run)
jiahaoLjh/PlaneSweepPose/lib/models/mvmppe.py
code served (permissive licence) · get_code("727d8abe1dd585d8")
How it ran

no invocation recorded.

Also present in other papers' code

Of 9 samples lifted from this paper's linked repositories, 9 carry a function identity and 0 do not. Of the 9 distinct functions with an identity, 0 are also present in other papers whose code we hold (0 with at least one paper dated earlier, 0 the same month, 0 later) and 9 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.

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

Existing approaches for multi-view multi-person 3D pose estimation explicitly establish cross-view correspondences to group 2D pose detections from multiple camera views and solve for the 3D pose estimation for each person. Establishing cross-view correspondences is challenging in multi-person scenes, and incorrect correspondences will lead to sub-optimal performance for the multi-stage pipeline. In this work, we present our multi-view 3D pose estimation approach based on plane sweep stereo to jointly address the cross-view fusion and 3D pose reconstruction in a single shot. Specifically, we propose to perform depth regression for each joint of each 2D pose in a target camera view. Cross-view consistency constraints are implicitly enforced by multiple reference camera views via the plane sweep algorithm to facilitate accurate depth regression. We adopt a coarse-to-fine scheme to first regress the person-level depth followed by a per-person joint-level relative depth estimation. 3D poses are obtained from a simple back-projection given the estimated depths. We evaluate our approach on benchmark datasets where it outperforms previous state-of-the-arts while being remarkably efficient. Our code is available at the project website. 1

For agents

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

get_harvested_code_for_paper("2104.02273")
get_code_for_paper("2104.02273")
have("2104.02273")

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

curl https://syntology.ai/api/ran/2104.02273.json

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

[![Syntology run record](https://syntology.ai/api/ran/2104.02273.svg)](https://syntology.ai/paper/2104.02273)

Connect an agent — have() is free.