SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 1904.04514 · 2019

High-Resolution Representations for Labeling Pixels and Regions

arXiv · PDF · Open in the Atlas

Code that ran

14 of 18 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 6 have no contract verdict; 8 failures of our instrument; 4 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/1904.04514.json · badge: /api/ran/1904.04514.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
kukby/Mish-semantic-segmentation-pytorch community (classifier label)
classifier basis: plan_row
6 ran · 0 no recorded run
liuch37/semantic-segmentation community (classifier label)
classifier basis not recorded on the link
3 ran · 1 no recorded run
mdt48/semantic-segmentation-pytorch community (classifier label)
classifier basis not recorded on the link
3 ran · 0 no recorded run
copy not recorded community
not in the paper's declared repositories
2 ran · 2 no recorded run
mohanliu/HRnet-semantic-segmentation-exp reimplementation (classifier label)
classifier basis: classifier_2025
0 ran · 1 no recorded run
FunctionStatusWhere it lives
as_variable Ran
ran on a synthesized input
liuch37/semantic-segmentation/mit_semseg/lib/utils/th.py
code served (permissive licence) · get_code("10bb6a2d70d61ca6")
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
import collections
from torch.autograd import Variable

torch.manual_seed(0)

# Test with a simple tensor
tensor_obj = torch.randn(2, 4)

# Test with a sequence (list)
sequence_obj = [torch.randn(2, 4), torch.randn(3, 5)]

# Test with a mapping (dict)
mapping_obj = {
    'a': torch.randn(2, 4),
    'b': torch.randn(3, 5)
}

# Test with nested structure
nested_obj = {
    'tensors': [torch.randn(2, 4), torch.randn(1, 3)],
    'single': torch.randn(4, 2)
}

# Create ARGS dict mapping the parameter 'obj' to each test case
ARGS = {
    'obj': tensor_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
calc_iou Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
this paper's copy was not recorded; identical code first harvested from HRNet/HRNet-MaskRCNN-Benchmark
pointer only · get_code("451bcfd505206dce")
How it ran

The body of this sample is withheld: this paper's copy of the code was not recorded (its link names no repository or source path), so no licence was read for it. 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.

conv3x3 Ran
ran; our contract draft was wrong, not the code
kukby/Mish-semantic-segmentation-pytorch/models/hrnet.py
pointer only (licence: BSD-3-Clause); 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 (BSD-3-Clause), 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.

crop Ran
ran on a synthesized input
mdt48/semantic-segmentation-pytorch/Colors/score.py
code served (permissive licence) · get_code("aae07a37335e79dd")
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 tempfile
import os

# Create a temporary image file for cv2.imread to read
temp_dir = tempfile.mkdtemp()
image_path = os.path.join(temp_dir, "test_image.png")

# Create a simple test image (100x200 RGB image)
test_image = np.random.randint(0, 256, (100, 200, 3), dtype=np.uint8)
cv2.imwrite(image_path, test_image)

ARGS = {
    "image": image_path
}
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
find Ran
ran on a synthesized input
mdt48/semantic-segmentation-pytorch/adjust_hotels.py
code served (permissive licence) · get_code("b674ceff022ebcc2")
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 int, ndim 0, dtype family int, 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)

search_term = "art/test"
ade = ["art", "test", "example"]

ARGS = {
    "search_term": search_term,
    "ade": ade
}
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 int, ndim 0, dtype family int, 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
find_recursive Ran
ran; our contract draft was wrong, not the code
kukby/Mish-semantic-segmentation-pytorch/utils.py
pointer only (licence: BSD-3-Clause); the stored copy get_code serves, from YuxiaoWang-AI/P3HOT, is pointer only (licence: NONE) · get_code("6890565633b480b8")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (BSD-3-Clause), but get_code serves the stored copy, first harvested from YuxiaoWang-AI/P3HOT, 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.

imresize Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
kukby/Mish-semantic-segmentation-pytorch/dataset.py
pointer only (licence: BSD-3-Clause); the stored copy get_code serves, from YuxiaoWang-AI/P3HOT, is pointer only (licence: NONE) · get_code("e23968c493407287")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (BSD-3-Clause), but get_code serves the stored copy, first harvested from YuxiaoWang-AI/P3HOT, 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.

load_url Ran
ran; our contract draft was wrong, not the code
kukby/Mish-semantic-segmentation-pytorch/models/utils.py
pointer only (licence: BSD-3-Clause); the stored copy get_code serves, from YuxiaoWang-AI/P3HOT, is pointer only (licence: NONE) · get_code("d087e320591e6ba7")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (BSD-3-Clause), but get_code serves the stored copy, first harvested from YuxiaoWang-AI/P3HOT, 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.

mark_volatile Ran
ran on a synthesized input
liuch37/semantic-segmentation/mit_semseg/lib/utils/th.py
code served (permissive licence) · get_code("32f45cf75891bd16")
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
from torch.autograd import Variable
import collections

torch.manual_seed(0)

# Test case 1: Tensor input
tensor_input = torch.randn(2, 4)

# Test case 2: Variable input
variable_input = Variable(torch.randn(2, 4))

# Test case 3: Dict/Mapping input
dict_input = {
    'tensor': torch.randn(2, 4),
    'variable': Variable(torch.randn(2, 4)),
    'nested': {
        'value': torch.randn(2, 4)
    }
}

# Test case 4: List/Sequence input
list_input = [
    torch.randn(2, 4),
    Variable(torch.randn(2, 4)),
    {'key': torch.randn(2, 4)}
]

# Test case 5: Scalar/other type
scalar_input = 42

# The function takes one parameter: obj
# We'll test with a tensor as the primary case
ARGS = {
    'obj': torch.randn(2, 4)
}
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
match Ran
ran on a synthesized input
mdt48/semantic-segmentation-pytorch/adjust_hotels.py
code served (permissive licence) · get_code("98a265524b160e10")
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

# Define the arguments for the match function
# Based on the function body, it takes:
# - hotels: an iterable of strings
# - ade: an iterable that supports 'in' checks and .index() method

hotels = ["art", "blue", "red", "green"]
ade = ["color1", "color2", "color3"]

ARGS = {
    "hotels": hotels,
    "ade": ade
}
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
round2nearest_multiple Ran
ran on a synthesized input
liuch37/semantic-segmentation/semantic_segmentation.py
code served (permissive licence) · get_code("cc83fb3836123035")
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 int, ndim 0, dtype family int, 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
import numpy as np

torch.manual_seed(0)

# Analyze the function: round2nearest_multiple(x, p)
# The function takes x and p, performs integer division and arithmetic operations
# x and p should be integers or integer-like values
# The function rounds x up to the nearest multiple of p

ARGS = {
    "x": 10,
    "p": 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 int, ndim 0, dtype family int, 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
setup_logger Ran
ran; our contract draft was wrong, not the code
kukby/Mish-semantic-segmentation-pytorch/utils.py
pointer only (licence: BSD-3-Clause); the stored copy get_code serves, from YuxiaoWang-AI/P3HOT, is pointer only (licence: NONE) · get_code("b50608a902a34b99")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (BSD-3-Clause), but get_code serves the stored copy, first harvested from YuxiaoWang-AI/P3HOT, 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.

unique Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
kukby/Mish-semantic-segmentation-pytorch/utils.py
pointer only (licence: BSD-3-Clause); the stored copy get_code serves, from YuxiaoWang-AI/P3HOT, is pointer only (licence: NONE) · get_code("5feabe579b35978a")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (BSD-3-Clause), but get_code serves the stored copy, first harvested from YuxiaoWang-AI/P3HOT, 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.

voting Ran
ran; our contract draft was wrong, not the code
this paper's copy was not recorded; identical code first harvested from HRNet/HRNet-MaskRCNN-Benchmark
pointer only · get_code("bb37605b18846a6b")
How it ran

The body of this sample is withheld: this paper's copy of the code was not recorded (its link names no repository or source path), so no licence was read for it. 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.

as_numpy Not yet run
unverified (harvested, no recorded run)
liuch37/semantic-segmentation/mit_semseg/lib/utils/th.py
code served (permissive licence) · get_code("c08e7121f699a983")
How it ran

no invocation recorded.

get_pose_net Not yet run
unverified (harvested, no recorded run)
this paper's copy was not recorded; identical code first harvested from wsjzha/deep-high-resolution-net.pytorch
pointer only · get_code("2534c81208f21998")
How it ran

The body of this sample is withheld: this paper's copy of the code was not recorded (its link names no repository or source path), so no licence was read for it. 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. No invocation recorded.

get_pose_net Not yet run
unverified (harvested, no recorded run)
this paper's copy was not recorded; identical code first harvested from visionNoob/hrnet_pytorch
pointer only · get_code("78015ae361e585ca")
How it ran

The body of this sample is withheld: this paper's copy of the code was not recorded (its link names no repository or source path), so no licence was read for it. 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. No invocation recorded.

get_seg_model Not yet run
unverified (harvested, no recorded run)
mohanliu/HRnet-semantic-segmentation-exp/lib/models/seg_hrnet.py
pointer only (licence: NOASSERTION) · get_code("7ae17b8a73133c4d")
How it ran

The body of this sample is withheld: GitHub could not classify the licence of this paper's repository, mohanliu/HRnet-semantic-segmentation-exp (NOASSERTION). 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. No invocation recorded.

Also present in other papers' code

Of 18 samples lifted from this paper's linked repositories, 18 carry a function identity and 0 do not. Of the 18 distinct functions with an identity, 15 are also present in other papers whose code we hold (14 with at least one paper dated earlier, 1 the same month, 12 later) and 3 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
102 dated earlier, 16 the same month, 568 later · 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
imresize 6 other papers
2 dated earlier, 4 later
1807.10221 (2018-07)
Unified Perceptual Parsing for Scene Understanding
1810.04020 (2018-10)
A Comprehensive Survey of Deep Learning for Image Captioning
1908.07919 (2019-08)
Deep High-Resolution Representation Learning for Visual Recognition
and 3 more not named here
unique 6 other papers
2 dated earlier, 4 later
1807.10221 (2018-07)
Unified Perceptual Parsing for Scene Understanding
1807.11590 (2018-07)
Acquisition of Localization Confidence for Accurate Object Detection
1908.07919 (2019-08)
Deep High-Resolution Representation Learning for Visual Recognition
and 3 more not named here
get_pose_net 5 other papers
1 dated earlier, 4 later
1902.09212 (2019-02)
Deep High-Resolution Representation Learning for Human Pose Estimation
1908.10357 (2019-08)
HigherHRNet: Scale-Aware Representation Learning for Bottom-Up Human Pose Estimation
1908.11542 (2019-08)
Satellite Pose Estimation with Deep Landmark Regression and Nonlinear Pose Refinement
and 2 more not named here
get_pose_net 5 other papers
1 dated earlier, 4 later
1902.09212 (2019-02)
Deep High-Resolution Representation Learning for Human Pose Estimation
1908.07919 (2019-08)
Deep High-Resolution Representation Learning for Visual Recognition
1908.10357 (2019-08)
HigherHRNet: Scale-Aware Representation Learning for Bottom-Up Human Pose Estimation
and 2 more not named here
load_url 4 other papers
2 dated earlier, 2 later
1807.10221 (2018-07)
Unified Perceptual Parsing for Scene Understanding
1807.11590 (2018-07)
Acquisition of Localization Confidence for Accurate Object Detection
1908.07919 (2019-08)
Deep High-Resolution Representation Learning for Visual Recognition
and 1 more not named here
find_recursive 3 other papers
1 dated earlier, 2 later
1807.10221 (2018-07)
Unified Perceptual Parsing for Scene Understanding
1908.07919 (2019-08)
Deep High-Resolution Representation Learning for Visual Recognition
2507.01630 (2025-07)
Prompt Guidance and Human Proximal Perception for HOT Prediction with Regional Joint Loss
get_seg_model 3 other papers
3 later
Fan_CADTransformer_Panoptic_Symbol_Spotting_Transformer_for_CAD_Drawings_CVPR_2022_paper (2022 (year only))
CADTransformer: Panoptic Symbol Spotting Transformer for CAD Drawings
2203.13815 (2022-03)
Versatile Multi-Modal Pre-Training for Human-Centric Perception
2303.04249 (2023-03)
Where We Are and What We're Looking At: Query Based Worldwide Image Geo-localization Using Hierarchies and Scenes
setup_logger 3 other papers
1 dated earlier, 2 later
1807.10221 (2018-07)
Unified Perceptual Parsing for Scene Understanding
1908.07919 (2019-08)
Deep High-Resolution Representation Learning for Visual Recognition
2507.01630 (2025-07)
Prompt Guidance and Human Proximal Perception for HOT Prediction with Regional Joint Loss
as_numpy 2 other papers
1 dated earlier, 1 later
1807.10221 (2018-07)
Unified Perceptual Parsing for Scene Understanding
1908.07919 (2019-08)
Deep High-Resolution Representation Learning for Visual Recognition
as_variable 2 other papers
1 dated earlier, 1 later
1807.10221 (2018-07)
Unified Perceptual Parsing for Scene Understanding
1908.07919 (2019-08)
Deep High-Resolution Representation Learning for Visual Recognition
mark_volatile 2 other papers
1 dated earlier, 1 later
1807.10221 (2018-07)
Unified Perceptual Parsing for Scene Understanding
1908.07919 (2019-08)
Deep High-Resolution Representation Learning for Visual Recognition
round2nearest_multiple 2 other papers
2 dated earlier
1807.10221 (2018-07)
Unified Perceptual Parsing for Scene Understanding
1807.11590 (2018-07)
Acquisition of Localization Confidence for Accurate Object Detection
calc_iou 1 other paper
1 dated earlier
1902.09212 (2019-02)
Deep High-Resolution Representation Learning for Human Pose Estimation
voting 1 other paper
1 dated earlier
1902.09212 (2019-02)
Deep High-Resolution Representation Learning for Human Pose Estimation

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

More are linked than shown here.

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

Abstract

High-resolution representation learning plays an essential role in many vision problems, e.g., pose estimation and semantic segmentation. The high-resolution network (HRNet)~\cite{SunXLW19}, recently developed for human pose estimation, maintains high-resolution representations through the whole process by connecting high-to-low resolution convolutions in \emph{parallel} and produces strong high-resolution representations by repeatedly conducting fusions across parallel convolutions. In this paper, we conduct a further study on high-resolution representations by introducing a simple yet effective modification and apply it to a wide range of vision tasks. We augment the high-resolution representation by aggregating the (upsampled) representations from all the parallel convolutions rather than only the representation from the high-resolution convolution as done in~\cite{SunXLW19}. This simple modification leads to stronger representations, evidenced by superior results. We show top results in semantic segmentation on Cityscapes, LIP, and PASCAL Context, and facial landmark detection on AFLW, COFW, $300$W, and WFLW. In addition, we build a multi-level representation from the high-resolution representation and apply it to the Faster R-CNN object detection framework and the extended frameworks. The proposed approach achieves superior results to existing single-model networks on COCO object detection. The code and models have been publicly available at \url{https://github.com/HRNet}.

For agents

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

get_harvested_code_for_paper("1904.04514")
get_code_for_paper("1904.04514")
have("1904.04514")

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

curl https://syntology.ai/api/ran/1904.04514.json

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

[![Syntology run record](https://syntology.ai/api/ran/1904.04514.svg)](https://syntology.ai/paper/1904.04514)

Connect an agent — have() is free.