SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2110.11219 · 2021

PlaneRecNet: Multi-Task Learning with Cross-Task Consistency for Piece-Wise Plane Detection and Reconstruction from a Single RGB Image

arXiv · PDF · Open in the Atlas

Code that ran

8 of 8 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 8 have no contract verdict; 0 failures of our instrument; 0 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/2110.11219.json · badge: /api/ran/2110.11219.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
eryixie/planerecnet canonical (classifier label)
classifier basis not recorded on the link
8 ran · 0 no recorded run
FunctionStatusWhere it lives
bbox_iou Ran
ran on a synthesized input
eryixie/planerecnet/models/functions/funcs.py
code served (permissive licence) · get_code("98ff672b9011e1de")
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; 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)

# bbox_iou takes two parameters: box_a and box_b
# Both should be tensors representing bounding boxes in format [x1, y1, x2, y2]
# The function handles both 2D and 3D inputs (with batch dimension)
# For testing, we'll use 2D inputs which will be converted to 3D internally

# Create ground truth bounding boxes: shape [num_objects, 4]
# Each box is [x1, y1, x2, y2]
box_a = torch.tensor([
    [10.0, 20.0, 50.0, 60.0],
    [15.0, 25.0, 55.0, 65.0]
], dtype=torch.float32)

# Create prior boxes: shape [num_priors, 4]
box_b = torch.tensor([
    [12.0, 22.0, 48.0, 62.0],
    [20.0, 30.0, 40.0, 50.0],
    [5.0, 15.0, 60.0, 70.0]
], dtype=torch.float32)

ARGS = {
    'box_a': box_a,
    'box_b': box_b
}
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 — no output shape is stated: observe_signature.py could not observe it (import:OSError)
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
compute_gradient_map Ran
ran on a synthesized input
eryixie/planerecnet/models/functions/losses.py
code served (permissive licence) · get_code("88364886ebfa0aee")
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. 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.functional as F

torch.manual_seed(0)

# Create a small depth map tensor
# Shape: (batch=2, channels=1, height=4, width=3)
depth_map = torch.randn(2, 1, 4, 3)

# Create a valid mask with the same spatial dimensions as the output
# After conv2d with 3x3 kernel and reflect padding on (4, 3) input,
# output will be (2, 1, 4, 3)
valid_mask = torch.ones(2, 1, 4, 3)

ARGS = {
    "depth_map": depth_map,
    "valid_mask": valid_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 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
construct_backbone Ran
ran on a synthesized input
eryixie/planerecnet/models/backbone.py
code served (permissive licence) · get_code("87523070b15ff05c")
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 SimpleBackbone. 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


class SimpleBackbone(nn.Module):
    def __init__(self, input_dim=4, hidden_dim=8):
        super().__init__()
        self.layers = nn.ModuleList()
        self.input_dim = input_dim
        self.hidden_dim = hidden_dim
        # Initialize with one layer
        self.layers.append(nn.Linear(input_dim, hidden_dim))
    
    def add_layer(self):
        """Add a new layer to the backbone."""
        self.layers.append(nn.Linear(self.hidden_dim, self.hidden_dim))
    
    def forward(self, x):
        for layer in self.layers:
            x = layer(x)
        return x


class BackboneConfig:
    def __init__(self, type_class, args, selected_layers):
        self.type = type_class
        self.args = args
        self.selected_layers = selected_layers


torch.manual_seed(0)

cfg = BackboneConfig(
    type_class=SimpleBackbone,
    args=(4, 8),  # input_dim, hidden_dim
    selected_layers=[0, 1, 2]  # Request layers 0, 1, 2 (so max is 2, need 3 layers total)
)

ARGS = {
    "cfg": cfg
}
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 SimpleBackbone — a SimpleBackbone output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a SimpleBackbone 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
intersect Ran
ran on a synthesized input
eryixie/planerecnet/models/functions/funcs.py
code served (permissive licence) · get_code("2b050648fe0422de")
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; 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)

# Create bounding box tensors
# Shape requirements: [n, A, 4] and [n, B, 4]
# where n=batch size, A=number of boxes in first set, B=number of boxes in second set
# Each box is [x1, y1, x2, y2]

n = 2  # batch size
A = 3  # number of boxes in box_a
B = 4  # number of boxes in box_b

# box_a: [n, A, 4] = [2, 3, 4]
box_a = torch.tensor([
    [[0.0, 0.0, 10.0, 10.0], [5.0, 5.0, 15.0, 15.0], [20.0, 20.0, 25.0, 25.0]],
    [[1.0, 1.0, 11.0, 11.0], [6.0, 6.0, 16.0, 16.0], [21.0, 21.0, 26.0, 26.0]]
], dtype=torch.float32)

# box_b: [n, B, 4] = [2, 4, 4]
box_b = torch.tensor([
    [[0.0, 0.0, 5.0, 5.0], [8.0, 8.0, 12.0, 12.0], [20.0, 20.0, 23.0, 23.0], [30.0, 30.0, 35.0, 35.0]],
    [[1.0, 1.0, 6.0, 6.0], [9.0, 9.0, 13.0, 13.0], [21.0, 21.0, 24.0, 24.0], [31.0, 31.0, 36.0, 36.0]]
], dtype=torch.float32)

ARGS = {
    "box_a": box_a,
    "box_b": box_b
}
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 — no output shape is stated: observe_signature.py could not observe it (import:OSError)
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
mask_iou Ran
ran on a synthesized input
eryixie/planerecnet/models/functions/funcs.py
code served (permissive licence) · get_code("e215e8d436fe5e29")
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. 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 two sets of masks
# masks_a: [a, h, w] where a=2, h=4, w=3
# masks_b: [b, h, w] where b=3, h=4, w=3
masks_a = torch.randint(0, 2, (2, 4, 3)).float()
masks_b = torch.randint(0, 2, (3, 4, 3)).float()

ARGS = {
    "masks_a": masks_a,
    "masks_b": masks_b,
    "iscrowd": False,
}
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
mask_nms Ran
ran on a synthesized input
eryixie/planerecnet/models/functions/nms.py
code served (permissive licence) · get_code("5abdae31a5987a25")
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 1, dtype family int, shape (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)

# Create small tensors for mask NMS
n_samples = 2
height, width = 4, 3

# cate_labels: tensor of class labels for each sample
cate_labels = torch.tensor([0, 0], dtype=torch.long)

# seg_masks: binary segmentation masks (n_samples, height, width)
seg_masks = torch.tensor([
    [[1, 0, 1],
     [1, 1, 0],
     [0, 1, 1],
     [1, 0, 0]],
    [[1, 1, 0],
     [0, 1, 1],
     [1, 0, 1],
     [0, 1, 0]]
], dtype=torch.uint8)

# sum_masks: pre-computed sum of pixels in each mask
sum_masks = torch.tensor([6, 6], dtype=torch.float32)

# cate_scores: confidence scores for each sample
cate_scores = torch.tensor([0.9, 0.8], dtype=torch.float32)

# nms_thr: IoU threshold for NMS
nms_thr = 0.5

ARGS = {
    "cate_labels": cate_labels,
    "seg_masks": seg_masks,
    "sum_masks": sum_masks,
    "cate_scores": cate_scores,
    "nms_thr": nms_thr
}
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 1, dtype family int, shape (2,) — shape (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
matrix_nms Ran
ran on a synthesized input
eryixie/planerecnet/models/functions/nms.py
code served (permissive licence) · get_code("b05bf3266d440dd9")
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 1, dtype family float, shape (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)

# Construct valid arguments for matrix_nms

# cate_labels: tensor of class labels for each detection
# Shape: (n_samples,) - we use n_samples=2
cate_labels = torch.tensor([0, 1], dtype=torch.long)

# seg_masks: segmentation masks for each detection
# Shape: (n_samples, height, width) - we use (2, 4, 3) for distinct dimensions
seg_masks = torch.rand(2, 4, 3)

# sum_masks: sum of each mask (used in IoU calculation)
# Shape: (n_samples,) - sum across spatial dimensions
sum_masks = seg_masks.reshape(2, -1).sum(dim=1)

# cate_scores: confidence scores for each detection
# Shape: (n_samples,)
cate_scores = torch.tensor([0.9, 0.8], dtype=torch.float32)

# sigma: hyperparameter for gaussian kernel
sigma = 2.0

# kernel: type of kernel ('gaussian' or 'linear')
kernel = 'gaussian'

ARGS = {
    'cate_labels': cate_labels,
    'seg_masks': seg_masks,
    'sum_masks': sum_masks,
    'cate_scores': cate_scores,
    'sigma': sigma,
    'kernel': kernel,
}
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 1, dtype family float, shape (2,) — shape (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
point_nms Ran
ran on a synthesized input
eryixie/planerecnet/models/functions/nms.py
code served (permissive licence) · get_code("564680ea0585a610")
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, 5) 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)

# Create a small heat map tensor
# Shape: (batch=2, channels=3, height=4, width=5)
heat = torch.randn(2, 3, 4, 5)

# Define kernel parameter
kernel = 2

# Construct the ARGS dictionary
ARGS = {
    "heat": heat,
    "kernel": kernel,
}
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, 5) — shape (2, 3, 4, 5) 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

Also present in other papers' code

Of 8 samples lifted from this paper's linked repositories, 8 carry a function identity and 0 do not. Of the 8 distinct functions with an identity, 2 are also present in other papers whose code we hold (2 with at least one paper dated earlier, 0 the same month, 0 later) and 6 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
construct_backbone 4 other papers
4 dated earlier
1904.02689 (2019-04)
YOLACT: Real-time Instance Segmentation
1911.09070 (2019-11)
EfficientDet: Scalable and Efficient Object Detection
1912.06218 (2019-12)
YOLACT++: Better Real-time Instance Segmentation
and 1 more not named here
intersect 2 other papers
2 dated earlier
1912.06218 (2019-12)
YOLACT++: Better Real-time Instance Segmentation
2005.09007 (2020-05)
U$^2$-Net: Going Deeper with Nested U-Structure for Salient Object Detection

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

Piece-wise 3D planar reconstruction provides holistic scene understanding of man-made environments, especially for indoor scenarios. Most recent approaches focused on improving the segmentation and reconstruction results by introducing advanced network architectures but overlooked the dual characteristics of piece-wise planes as objects and geometric models. Different from other existing approaches, we start from enforcing cross-task consistency for our multi-task convolutional neural network, PlaneRecNet, which integrates a single-stage instance segmentation network for piece-wise planar segmentation and a depth decoder to reconstruct the scene from a single RGB image. To achieve this, we introduce several novel loss functions (geometric constraint) that jointly improve the accuracy of piece-wise planar segmentation and depth estimation. Meanwhile, a novel Plane Prior Attention module is used to guide depth estimation with the awareness of plane instances. Exhaustive experiments are conducted in this work to validate the effectiveness and efficiency of our method.

For agents

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

get_harvested_code_for_paper("2110.11219")
get_code_for_paper("2110.11219")
have("2110.11219")

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

curl https://syntology.ai/api/ran/2110.11219.json

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

[![Syntology run record](https://syntology.ai/api/ran/2110.11219.svg)](https://syntology.ai/paper/2110.11219)

Connect an agent — have() is free.