SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2205.13490 · CVPR · 2022

SemAffiNet: Semantic-Affine Transformation for Point Cloud Segmentation

Ziyi Wang, Yongming Rao, Jiwen Lu, Jie Zhou, Xumin Yu

arXiv · PDF · Open in the Atlas

Code that ran

11 of 11 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; 2 honoured the contract we drafted, 0 violated it and 5 have no contract verdict; 4 failures of our instrument; 0 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/2205.13490.json · badge: /api/ran/2205.13490.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
wangzy22/SemAffiNet canonical (classifier label)
classifier basis not recorded on the link
11 ran · 0 no recorded run
FunctionStatusWhere it lives
adjust_intrinsic Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
wangzy22/SemAffiNet/dataset/scanNetCross.py
code served (permissive licence) · get_code("81a52f12291c03c6")
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 (3, 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 math
import numpy as np

torch.manual_seed(0)

# Analyze the function:
# - intrinsic: a 2D array/tensor that gets indexed with [0, 0], [1, 1], [0, 2], [1, 2]
#   and elements are multiplied. This is a camera intrinsic matrix (typically 3x3).
# - intrinsic_image_dim: tuple/list with 2 elements (height, width of original image)
# - image_dim: tuple/list with 2 elements (height, width of target image)

# Create a sample camera intrinsic matrix (3x3)
intrinsic = torch.tensor([
    [500.0, 0.0, 320.0],
    [0.0, 500.0, 240.0],
    [0.0, 0.0, 1.0]
], dtype=torch.float32)

# Original image dimensions (height, width)
intrinsic_image_dim = (480, 640)

# Target image dimensions (height, width) - different from original
image_dim = (240, 320)

ARGS = {
    "intrinsic": intrinsic,
    "intrinsic_image_dim": intrinsic_image_dim,
    "image_dim": image_dim,
}
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 2, dtype family float, shape (3, 3) — shape (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
collation_fn Ran
ran on a synthesized input
wangzy22/SemAffiNet/dataset/scanNet3D.py
code served (permissive licence) · get_code("0f0c2365c06796db")
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)

# The function expects a batch parameter which is an iterable of tuples
# Each tuple contains (coords, feats, labels)
# coords should be a tensor that can be indexed with [:, 0]
# feats and labels should be tensors that can be concatenated

# Create sample data items
item1_coords = torch.randn(4, 4)  # 4 points, 4 features (x,y,z,batch)
item1_feats = torch.randn(4, 3)   # 4 points, 3 feature dims
item1_labels = torch.randint(0, 2, (4,))  # 4 labels

item2_coords = torch.randn(5, 4)  # 5 points, 4 features
item2_feats = torch.randn(5, 3)   # 5 points, 3 feature dims
item2_labels = torch.randint(0, 2, (5,))  # 5 labels

# Create the batch as a list of tuples
batch = [
    (item1_coords, item1_feats, item1_labels),
    (item2_coords, item2_feats, item2_labels),
]

ARGS = {
    "batch": batch
}
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
collation_fn Ran
ran on a synthesized input
wangzy22/SemAffiNet/dataset/scanNetCross.py
code served (permissive licence) · get_code("cb93a4455661b639")
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 a single batch item with the required components
# coords: N x 4 (batch, x, y, z)
coords_item = torch.randn(5, 4)

# feats: N x 3
feats_item = torch.randn(5, 3)

# labels: N
labels_item = torch.randint(0, 10, (5,))

# colors: B x C x H x W x V
# For a single item in batch, we need shape that will stack to B x C x H x W x V
colors_item = torch.randn(3, 8, 7, 2)

# labels_2d: B x H x W x V
# For a single item, shape that will stack to B x H x W x V
labels_2d_item = torch.randint(0, 10, (8, 7, 2))

# links: N x 4 x V
links_item = torch.randn(5, 4, 2)

# Create batch with 2 items
batch_item_1 = (coords_item, feats_item, labels_item, colors_item, labels_2d_item, links_item)

coords_item_2 = torch.randn(6, 4)
feats_item_2 = torch.randn(6, 3)
labels_item_2 = torch.randint(0, 10, (6,))
colors_item_2 = torch.randn(3, 8, 7, 2)
labels_2d_item_2 = torch.randint(0, 10, (8, 7, 2))
links_item_2 = torch.randn(6, 4, 2)

batch_item_2 = (coords_item_2, feats_item_2, labels_item_2, colors_item_2, labels_2d_item_2, links_item_2)

batch = [batch_item_1, batch_item_2]

ARGS = {
    "batch": batch
}
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
collation_fn_eval_all Ran
ran on a synthesized input
wangzy22/SemAffiNet/dataset/scanNet3D.py
code served (permissive licence) · get_code("50ae10fb57320b06")
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 sample batch items
# Each item is a tuple of (coords, feats, labels, inds_recons)

# Item 1
coords_1 = torch.randn(3, 4)  # 3 points, 4 coordinates (x,y,z,batch)
feats_1 = torch.randn(3, 8)   # 3 points, 8 features
labels_1 = torch.randint(0, 2, (3,))  # 3 labels
inds_recons_1 = torch.tensor([0, 1, 2])  # 3 reconstruction indices

# Item 2
coords_2 = torch.randn(5, 4)  # 5 points, 4 coordinates
feats_2 = torch.randn(5, 8)   # 5 points, 8 features
labels_2 = torch.randint(0, 2, (5,))  # 5 labels
inds_recons_2 = torch.tensor([0, 1, 2, 3, 4])  # 5 reconstruction indices

# Create batch as list of tuples
batch = [
    (coords_1, feats_1, labels_1, inds_recons_1),
    (coords_2, feats_2, labels_2, inds_recons_2)
]

ARGS = {
    'batch': batch
}
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
downsample_avg Ran
ran; our contract draft was wrong, not the code
wangzy22/SemAffiNet/models/resnet_d.py
code served (permissive licence) · get_code("6a8bcd1ebef3b13f")
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
from timm.models.layers import AvgPool2dSame

torch.manual_seed(0)

ARGS = {
    'in_channels': 3,
    'out_channels': 64,
    'kernel_size': 3,
    'stride': 2,
    'dilation': 1,
    'first_dilation': None,
    'norm_layer': nn.BatchNorm2d,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
timm, 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
downsample_conv Ran
ran; our contract draft was wrong, not the code
wangzy22/SemAffiNet/models/resnet_d.py
code served (permissive licence) · get_code("eea0376ea397a1c7")
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)

# Define ARGS dict with all parameters for downsample_conv
ARGS = {
    'in_channels': 3,
    'out_channels': 64,
    'kernel_size': 3,
    'stride': 2,
    'dilation': 1,
    'first_dilation': None,
    'norm_layer': nn.BatchNorm2d
}
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
get_coords_map Ran
ran on a synthesized input
wangzy22/SemAffiNet/models/bpm.py
code served (permissive licence) · get_code("ff48686fd2773fe0")
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 dict. 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)

# Stand-in classes for MinkowskiEngine-like coordinate system
class CoordinateMap:
    def __init__(self, key):
        self.key = key

class CoordinateManager:
    def __init__(self):
        self.stride_maps = {}
    
    def stride_map(self, key1, key2):
        # Return a mapping between two coordinate map keys
        return {"from": key1, "to": key2, "stride": 2}

class SparseTensor:
    def __init__(self, coordinate_manager, coordinate_map_key):
        self.coordinate_manager = coordinate_manager
        self.coordinate_map_key = coordinate_map_key

# Create shared coordinate manager
coord_manager = CoordinateManager()

# Create two sparse tensors with the same coordinate manager
x = SparseTensor(
    coordinate_manager=coord_manager,
    coordinate_map_key="layer_0"
)

y = SparseTensor(
    coordinate_manager=coord_manager,
    coordinate_map_key="layer_1"
)

ARGS = {
    "x": x,
    "y": y
}
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 dict — a dict output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a dict 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
get_padding Ran
ran, honoured the contract we drafted
wangzy22/SemAffiNet/models/resnet_d.py
pointer only (licence: MIT); the stored copy get_code serves, from xslidi/EfficientNets_ddl_apex, is pointer only (licence: NONE) · get_code("9b02c1068795fd5b")
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 xslidi/EfficientNets_ddl_apex, 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_stats Ran
ran on a synthesized input
wangzy22/SemAffiNet/dataset/pregroup_2d_scannet.py
code served (permissive licence) · get_code("b309b945f3526f50")
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 dict. 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 numpy as np

torch.manual_seed(0)

class MockGroup:
    def __init__(self):
        self.data = torch.randn(2, 4)
    
    def sum(self):
        return self.data.sum()

group = MockGroup()

ARGS = {
    'group': group
}
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 dict — a dict output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a dict 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
make_intrinsic Ran
ran, honoured the contract we drafted
wangzy22/SemAffiNet/dataset/scanNetCross.py
code served (permissive licence) · get_code("f47c8e54943ce833")
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 float, shape (4, 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 numpy as np

# Construct valid arguments for make_intrinsic
# The function takes 4 scalar parameters: fx, fy, mx, my (focal lengths and principal point)

ARGS = {
    "fx": 500.0,
    "fy": 500.0,
    "mx": 320.0,
    "my": 240.0,
}
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 float, shape (4, 4) — shape (4, 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
state_dict_remove_moudle Ran
ran; our contract draft was wrong, not the code
wangzy22/SemAffiNet/models/semaffinet.py
pointer only (licence: MIT); the stored copy get_code serves, from Wang-pengfei/GGSD, is pointer only (licence: NONE) · get_code("c4b88bc6388a332e")
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 Wang-pengfei/GGSD, 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.

Also present in other papers' code

Of 11 samples lifted from this paper's linked repositories, 11 carry a function identity and 0 do not. Of the 11 distinct functions with an identity, 6 are also present in other papers whose code we hold (3 with at least one paper dated earlier, 0 the same month, 6 later) and 5 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
get_padding 10 other papers
3 dated earlier, 7 later
1512.03385 (2015-12)
Deep Residual Learning for Image Recognition
2110.07641 (2021-10)
Non-deep Networks
2203.12827 (2022-03)
Sparse Instance Activation for Real-Time Instance Segmentation
and 7 more not named here
downsample_conv 7 other papers
2 dated earlier, 5 later
2110.07641 (2021-10)
Non-deep Networks
2203.12827 (2022-03)
Sparse Instance Activation for Real-Time Instance Segmentation
2208.10169 (2022-08)
Multi-Granularity Distillation Scheme Towards Lightweight Semi-Supervised Semantic Segmentation
and 4 more not named here
downsample_avg 6 other papers
2 dated earlier, 4 later
2110.07641 (2021-10)
Non-deep Networks
2203.12827 (2022-03)
Sparse Instance Activation for Real-Time Instance Segmentation
2208.10169 (2022-08)
Multi-Granularity Distillation Scheme Towards Lightweight Semi-Supervised Semantic Segmentation
and 3 more not named here
adjust_intrinsic 4 other papers
4 later
2311.17707 (2023-11)
SAMPro3D: Locating SAM Prompts in 3D for Zero-Shot Instance Segmentation
2312.10671 (2023-12)
Open3DIS: Open-Vocabulary 3D Instance Segmentation with 2D Mask Guidance
2409.03757 (2024-09)
Lexicon3D: Probing Visual Foundation Models for Complex 3D Scene Understanding
and 1 more not named here
make_intrinsic 4 other papers
4 later
2311.17707 (2023-11)
SAMPro3D: Locating SAM Prompts in 3D for Zero-Shot Instance Segmentation
2312.10671 (2023-12)
Open3DIS: Open-Vocabulary 3D Instance Segmentation with 2D Mask Guidance
2409.03757 (2024-09)
Lexicon3D: Probing Visual Foundation Models for Complex 3D Scene Understanding
and 1 more not named here
state_dict_remove_moudle 3 other papers
3 later
2211.15654 (2022-11)
OpenScene: 3D Scene Understanding with Open Vocabularies
2407.13362 (2024-07)
Open Vocabulary 3D Scene Understanding via Geometry Guided Self-Distillation
2503.16707 (2025-03)
Cross-Modal and Uncertainty-Aware Agglomeration for Open-Vocabulary 3D Scene Understanding

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

Conventional point cloud semantic segmentation methods usually employ an encoder-decoder architecture, where mid-level features are locally aggregated to extract geometric information. However, the over-reliance on these classagnostic local geometric representations may raise confusion between local parts from different categories that are similar in appearance or spatially adjacent. To address this issue, we argue that mid-level features can be further enhanced with semantic information, and propose semanticaffine transformation that transforms features of mid-level points belonging to different categories with class-specific affine parameters. Based on this technique, we propose Se-mAffiNet for point cloud semantic segmentation, which utilizes the attention mechanism in the Transformer module to implicitly and explicitly capture global structural knowledge within local parts for overall comprehension of each category. We conduct extensive experiments on the Scan-NetV2 and NYUv2 datasets, and evaluate semantic-affine transformation on various 3D point cloud and 2D image segmentation baselines, where both qualitative and quantitative results demonstrate the superiority and generalization ability of our proposed approach. Code is available at https://github.com/wangzy22/SemAffiNet.

For agents

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

get_harvested_code_for_paper("2205.13490")
get_code_for_paper("2205.13490")
have("2205.13490")

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

curl https://syntology.ai/api/ran/2205.13490.json

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

[![Syntology run record](https://syntology.ai/api/ran/2205.13490.svg)](https://syntology.ai/paper/2205.13490)

Connect an agent — have() is free.