SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2007.09365 · 2020

Malleable 2.5D Convolution: Learning Receptive Fields along the Depth-axis for RGB-D Scene Parsing

arXiv · PDF · Open in the Atlas

Code that ran

9 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, 1 violated it and 5 have no contract verdict; 3 failures of our instrument; 5 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/2007.09365.json · badge: /api/ran/2007.09365.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
charlesCXK/RGBD_Semantic_Segmentation_PyTorch canonical (classifier label)
classifier basis: census
5 ran · 0 no recorded run
David-zaiwang/114_rgbd_seg community (classifier label)
classifier basis not recorded on the link
4 ran · 5 no recorded run
FunctionStatusWhere it lives
compute_score Ran
ran on a synthesized input
David-zaiwang/114_rgbd_seg/furnace/seg_opr/metric.py
code served (permissive licence) · get_code("3e98ad928a86de29")
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

# Analyze the function:
# - hist: used with np.diag(), .sum(1), .sum(0), indexing with [1:]
#   This is a 2D confusion matrix array
# - correct: used in division, appears to be a scalar
# - labeled: used in division, appears to be a scalar

# Create a small confusion matrix (4x4)
# Rows = true classes, Cols = predicted classes
hist = np.array([
    [50, 5, 2, 1],
    [3, 40, 4, 2],
    [2, 3, 35, 5],
    [1, 2, 3, 30]
], dtype=np.float64)

# correct: number of correctly classified pixels
correct = 155

# labeled: total number of labeled pixels
labeled = 192

ARGS = {
    "hist": hist,
    "correct": correct,
    "labeled": labeled
}
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
conv1x1 Ran
ran; our contract draft was wrong, not the code
charlesCXK/RGBD_Semantic_Segmentation_PyTorch/model/SA-Gate.nyu.432/net_util.py
code served (permissive licence) · get_code("d9def42110729a85")
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 Conv2d. 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)

# The function conv1x1 takes in_planes, out_planes, and stride as parameters
# It returns an nn.Conv2d layer configured for 1x1 convolution
# We need to provide valid integer arguments for these parameters

ARGS = {
    "in_planes": 3,
    "out_planes": 64,
    "stride": 1
}
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 Conv2d — a Conv2d output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a Conv2d 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
conv3x3 Ran
ran; our contract draft was wrong, not the code
charlesCXK/RGBD_Semantic_Segmentation_PyTorch/furnace/base_model/resnet.py
pointer only (licence: MIT); the stored copy get_code serves, from lixiaotong97/DSU, is pointer only (licence: NONE) · get_code("fac5364e2f53c6db")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (MIT), but get_code serves the stored copy, first harvested from lixiaotong97/DSU, and GitHub detected no licence in that repository when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

get_a_var Ran
ran on a synthesized input
David-zaiwang/114_rgbd_seg/furnace/seg_opr/parallel/parallel_apply.py
code served (permissive licence) · get_code("74bcab5cc9c8e91b")
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 get_a_var takes a single parameter 'obj' and recursively searches
# for and returns the first torch.Tensor it finds within that object.
# It handles Tensors, lists, tuples, and dicts.

# Let's create a test object that contains tensors nested in various ways
tensor1 = torch.randn(2, 4)
tensor2 = torch.randn(3, 5)

# Create a complex nested structure with tensors
obj = {
    "model_weights": [tensor1, tensor2],
    "config": {
        "hidden_dim": 64,
        "output": (tensor2, "some_string")
    },
    "data": [1, 2, 3]
}

ARGS = {
    "obj": obj
}
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
hist_info Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
David-zaiwang/114_rgbd_seg/furnace/seg_opr/metric.py
pointer only (licence: MIT); the stored copy get_code serves, from harshm121/m3l, is pointer only (licence: NONE) · get_code("9e85fa995eda358a")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (MIT), but get_code serves the stored copy, first harvested from harshm121/m3l, and GitHub detected no licence in that repository when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

meanIoU Ran
ran on a synthesized input
David-zaiwang/114_rgbd_seg/furnace/seg_opr/metric.py
code served (permissive licence) · get_code("be62e3869efd8038")
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

# Construct valid arguments for meanIoU function
# The function expects:
# - area_intersection: array where np.sum(axis=1) is called
# - area_union: array where np.sum(axis=1) is called
# Both should have shape (num_classes, num_samples) or similar 2D structure

# Create small sample data
# Shape: (4, 3) - 4 classes, 3 samples
area_intersection = np.array([
    [10.0, 15.0, 12.0],
    [8.0, 12.0, 10.0],
    [5.0, 7.0, 6.0],
    [3.0, 4.0, 5.0]
], dtype=np.float32)

area_union = np.array([
    [20.0, 30.0, 25.0],
    [16.0, 24.0, 20.0],
    [10.0, 14.0, 12.0],
    [6.0, 8.0, 10.0]
], dtype=np.float32)

ARGS = {
    "area_intersection": area_intersection,
    "area_union": area_union
}
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
normalize Ran
ran, violated the contract we drafted
charlesCXK/RGBD_Semantic_Segmentation_PyTorch/model/SA-Gate.nyu.432/dataloader.py
pointer only (licence: MIT); the stored copy get_code serves, from harshm121/m3l, is pointer only (licence: NONE) · get_code("17cb5b0dfa01b491")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (MIT), but get_code serves the stored copy, first harvested from harshm121/m3l, and GitHub detected no licence in that repository when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

random_mirror Ran
ran on a synthesized input
charlesCXK/RGBD_Semantic_Segmentation_PyTorch/model/SA-Gate.nyu.432/dataloader.py
code served (permissive licence) · get_code("6e022fd18a9397fa")
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 cv2
import numpy as np
import random

# Set seed for reproducibility
random.seed(0)
np.random.seed(0)

# Create small test images using numpy (cv2 works with numpy arrays)
# Using distinct dimensions for clarity: height=4, width=3
height, width = 4, 3
channels = 3

# Create img: RGB image with shape (4, 3, 3)
img = np.random.randint(0, 256, (height, width, channels), dtype=np.uint8)

# Create gt: ground truth, typically grayscale or single channel with shape (4, 3, 1)
gt = np.random.randint(0, 2, (height, width, 1), dtype=np.uint8)

# Create hha: HHA (horizontal and height angle) representation with shape (4, 3, 3)
hha = np.random.randint(0, 256, (height, width, channels), dtype=np.uint8)

ARGS = {
    'img': img,
    'gt': gt,
    'hha': hha
}
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 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
random_scale Ran
ran on a synthesized input
charlesCXK/RGBD_Semantic_Segmentation_PyTorch/model/SA-Gate.nyu.432/dataloader.py
code served (permissive licence) · get_code("eda7f5af7267de5d")
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 cv2
import random
import numpy as np

random.seed(0)

# Create small test images using numpy
# Using uint8 for img and hha (typical for cv2 images)
# Using int32 or uint8 for gt (ground truth/segmentation mask)
height, width = 32, 48
img = np.random.randint(0, 256, (height, width, 3), dtype=np.uint8)
gt = np.random.randint(0, 10, (height, width), dtype=np.int32)
hha = np.random.randint(0, 256, (height, width, 3), dtype=np.uint8)

scales = [0.5, 1.0, 1.5, 2.0]

ARGS = {
    "img": img,
    "gt": gt,
    "hha": hha,
    "scales": scales,
}
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 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
batchnormtrain Not yet run
unverified (harvested, no recorded run)
David-zaiwang/114_rgbd_seg/furnace/seg_opr/sync_bn/functions.py
code served (permissive licence) · get_code("5c906904c2492dc7")
How it ran

no invocation recorded.

parallel_apply Not yet run
unverified (harvested, no recorded run)
David-zaiwang/114_rgbd_seg/furnace/seg_opr/parallel/parallel_apply.py
code served (permissive licence) · get_code("c490830f0e503906")
How it ran

no invocation recorded.

parallel_apply Not yet run
unverified (harvested, no recorded run)
David-zaiwang/114_rgbd_seg/furnace/seg_opr/sync_bn/parallel_apply.py
code served (permissive licence) · get_code("58954eadae7321d1")
How it ran

no invocation recorded.

patch_replication_callback Not yet run
unverified (harvested, no recorded run)
David-zaiwang/114_rgbd_seg/furnace/seg_opr/sync_bn/parallel.py
code served (permissive licence) · get_code("27e207fe8a74c3e7")
How it ran

no invocation recorded.

sum_square Not yet run
unverified (harvested, no recorded run)
David-zaiwang/114_rgbd_seg/furnace/seg_opr/sync_bn/functions.py
code served (permissive licence) · get_code("8a79f49da87d44bf")
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, 8 are also present in other papers whose code we hold (6 with at least one paper dated earlier, 2 the same month, 6 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
conv3x3 686 other papers
238 dated earlier, 16 the same month, 431 later, 1 same year, order not determinable · 31 only as an assembly brick
1406.2661 (2014-06)
Generative Adversarial Networks
1409.0575 (2014-09)
ImageNet Large Scale Visual Recognition Challenge
1412.6572 (2014-12)
Explaining and Harnessing Adversarial Examples
and 683 more not named here
conv1x1 431 other papers
83 dated earlier, 9 the same month, 338 later, 1 undated · 38 only as an assembly brick
1409.1556 (2014-09)
Very Deep Convolutional Networks for Large-Scale Image Recognition
1506.01186 (2015-06)
Cyclical Learning Rates for Training Neural Networks
1507.05717 (2015-07)
An End-to-End Trainable Neural Network for Image-based Sequence Recognition and Its Application to Scene Text Recognition
and 428 more not named here
patch_replication_callback 16 other papers
8 dated earlier, 8 later
1612.01105 (2016-12)
Pyramid Scene Parsing Network
1807.10221 (2018-07)
Unified Perceptual Parsing for Scene Understanding
1904.01941 (2019-04)
Character Region Awareness for Text Detection
and 13 more not named here
hist_info 3 other papers
2 dated earlier, 1 later
1808.00897 (2018-08)
BiSeNet: Bilateral Segmentation Network for Real-time Semantic Segmentation
1909.12513 (2019-09)
Learnable Tree Filter for Structure-preserving Feature Transform
2106.01226 (2021-06)
Semi-Supervised Semantic Segmentation with Cross Pseudo Supervision
normalize 3 other papers
3 later
2106.01226 (2021-06)
Semi-Supervised Semantic Segmentation with Cross Pseudo Supervision
2208.10169 (2022-08)
Multi-Granularity Distillation Scheme Towards Lightweight Semi-Supervised Semantic Segmentation
2411.03829 (2024-11)
Generalize or Detect? Towards Robust Semantic Segmentation Under Multiple Distribution Shifts
compute_score 2 other papers
2 dated earlier
1808.00897 (2018-08)
BiSeNet: Bilateral Segmentation Network for Real-time Semantic Segmentation
1909.12513 (2019-09)
Learnable Tree Filter for Structure-preserving Feature Transform
meanIoU 2 other papers
2 dated earlier
1808.00897 (2018-08)
BiSeNet: Bilateral Segmentation Network for Real-time Semantic Segmentation
1909.12513 (2019-09)
Learnable Tree Filter for Structure-preserving Feature Transform
get_a_var 1 other paper
1 later
2021.findings-emnlp.386 (2021 (year only))
Glyph Enhanced Chinese Character Pre-Training for Lexical Sememe Prediction

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

Depth data provide geometric information that can bring progress in RGB-D scene parsing tasks. Several recent works propose RGB-D convolution operators that construct receptive fields along the depth-axis to handle 3D neighborhood relations between pixels. However, these methods pre-define depth receptive fields by hyperparameters, making them rely on parameter selection. In this paper, we propose a novel operator called malleable 2.5D convolution to learn the receptive field along the depth-axis. A malleable 2.5D convolution has one or more 2D convolution kernels. Our method assigns each pixel to one of the kernels or none of them according to their relative depth differences, and the assigning process is formulated as a differentiable form so that it can be learnt by gradient descent. The proposed operator runs on standard 2D feature maps and can be seamlessly incorporated into pre-trained CNNs. We conduct extensive experiments on two challenging RGB-D semantic segmentation dataset NYUDv2 and Cityscapes to validate the effectiveness and the generalization ability of our method.

For agents

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

get_harvested_code_for_paper("2007.09365")
get_code_for_paper("2007.09365")
have("2007.09365")

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

curl https://syntology.ai/api/ran/2007.09365.json

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

[![Syntology run record](https://syntology.ai/api/ran/2007.09365.svg)](https://syntology.ai/paper/2007.09365)

Connect an agent — have() is free.