SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2304.07773 · 2023

PCPNet: An Efficient and Semantic-Enhanced Transformer Network for Point Cloud Prediction

arXiv · PDF · Open in the Atlas

Code that ran

7 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 7 have no contract verdict; 0 failures of our instrument; 1 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/2304.07773.json · badge: /api/ran/2304.07773.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
blurryface0814/pcpnet canonical (classifier label)
classifier basis: census
7 ran · 1 no recorded run
FunctionStatusWhere it lives
get_filename Ran
ran on a synthesized input
blurryface0814/pcpnet/pcpnet/utils/visualization.py
code served (permissive licence) · get_code("7d19748285a7244b")
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 tempfile
import os
import glob

# Create a temporary directory with some dummy .ply files
temp_dir = tempfile.mkdtemp()

# Create dummy .ply files with numeric names
for i in range(5):
    filename = os.path.join(temp_dir, f"{i}.ply")
    with open(filename, 'w') as f:
        f.write("dummy ply content")

# Ensure the path ends with '/' for the glob pattern
path = temp_dir + '/'

# Create a simple index to retrieve
idx = 2

ARGS = {
    'path': path,
    'idx': idx
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
none outside the standard library — 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
get_pred_and_gt_point_cloud Ran
ran on a synthesized input
blurryface0814/pcpnet/pcpnet/utils/logger.py
code served (permissive licence) · get_code("11b1310e42fc2b66")
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
import torch.nn as nn

torch.manual_seed(0)

class MockProjection:
    """Mock projection class that implements the methods used in the function"""
    
    def get_masked_range_view(self, output):
        # Should return a tensor of shape [batch_size, steps, height, width]
        # Used as: masked_prediction[sample_index, step, :, :]
        return torch.rand(2, 5, 4, 4)
    
    def get_valid_points_from_range_view(self, range_view):
        # Takes a [height, width] range view and returns valid points
        # Should return shape that can be reshaped to [1, -1, 3]
        # range_view is [4, 4], extract some valid points
        mask = range_view > 0.5
        num_points = mask.sum().item()
        if num_points == 0:
            num_points = 5
        return torch.rand(num_points, 3)

# Create projection instance
projection = MockProjection()

# Create batch dict
# batch['fut_data'] should be indexable as [sample_index, channel, step, height, width]
# sample_index=0, channels need: [0] for range, [1:4] for points (x,y,z)
# step ranges, height and width for spatial dimensions
batch = {
    'fut_data': torch.rand(2, 4, 5, 4, 4),  # [batch=2, channels=4, steps=5, height=4, width=4]
    'past_data': torch.rand(2, 4, 3, 4, 4),
}

# Create output dict
output = {
    'predicted_logits': torch.randn(2, 5, 4, 4),
}

# Define other parameters
step = 2
sample_index = 0

ARGS = {
    'projection': projection,
    'batch': batch,
    'output': output,
    'step': step,
    'sample_index': sample_index,
}
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
last_file Ran
ran on a synthesized input
blurryface0814/pcpnet/pcpnet/utils/visualization.py
code served (permissive licence) · get_code("f5b9b639e1f32818")
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 os
import tempfile
import glob

# Create a temporary directory with mock .ply files
temp_dir = tempfile.mkdtemp()

# Create some dummy .ply files with numeric names
for i in range(5):
    filename = os.path.join(temp_dir, f"{i}.ply")
    with open(filename, 'w') as f:
        f.write("ply\n")

# The path should end with '/' for the glob pattern to work correctly
path = temp_dir + '/'

ARGS = {
    'path': path
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
none outside the standard library — 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
load_calib Ran
ran on a synthesized input
blurryface0814/pcpnet/pcpnet/utils/utils.py
code served (permissive licence) · get_code("f32f9f6ed893dfc6")
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
import tempfile
import os

# Create a temporary calibration file with valid content
temp_dir = tempfile.mkdtemp()
calib_file = os.path.join(temp_dir, "calib.txt")

# Write calibration data in KITTI format
calib_content = """P0: 1 2 3 4 5 6 7 8 9 10 11 12
P1: 1 2 3 4 5 6 7 8 9 10 11 12
P2: 1 2 3 4 5 6 7 8 9 10 11 12
P3: 1 2 3 4 5 6 7 8 9 10 11 12
R0_rect: 1 0 0 0 1 0 0 0 1
Tr: 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 1.1 1.2
Tr_imu_to_velo: 1 0 0 0 0 1 0 0 0 0 1 0 0 0 0 1
"""

with open(calib_file, 'w') as f:
    f.write(calib_content)

ARGS = {
    'calib_path': calib_file
}
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
load_poses Ran
ran on a synthesized input
blurryface0814/pcpnet/pcpnet/utils/utils.py
code served (permissive licence) · get_code("ad221712326a962d")
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 3, dtype family float, shape (2, 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
import tempfile
import os

# Create a temporary directory for test files
temp_dir = tempfile.mkdtemp()

# Create a test .txt pose file with 2 poses (3x4 transformation matrices per pose)
txt_pose_file = os.path.join(temp_dir, 'poses.txt')
with open(txt_pose_file, 'w') as f:
    # First pose: 3x4 matrix flattened to 12 values
    f.write('1.0 0.0 0.0 0.5 0.0 1.0 0.0 1.5 0.0 0.0 1.0 2.5\n')
    # Second pose: another 3x4 matrix
    f.write('0.9 0.1 0.0 0.6 -0.1 0.9 0.0 1.6 0.0 0.0 1.0 2.6\n')

ARGS = {
    'pose_path': txt_pose_file
}
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 3, dtype family float, shape (2, 4, 4) — shape (2, 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
make_path Ran
ran on a synthesized input
blurryface0814/pcpnet/pcpnet/utils/logger.py
code served (permissive licence) · get_code("91f79d76865ecfcf")
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 str. 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 os
import tempfile

# Create a temporary directory for testing
test_dir = tempfile.mkdtemp()
test_path = os.path.join(test_dir, "new_directory")

# Construct arguments for make_path function
ARGS = {
    "path": test_path
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
none outside the standard library — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type str — a str output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a str 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
range_projection Ran
ran on a synthesized input
blurryface0814/pcpnet/pcpnet/utils/utils.py
code served (permissive licence) · get_code("96249332b0ddac61")
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

# Create a small pointcloud with 10 points
# Each point has [x, y, z, intensity]
np.random.seed(0)
num_points = 10
current_vertex = np.random.randn(num_points, 4).astype(np.float32)
# Ensure points are not too far from origin for realistic range values
current_vertex[:, :3] *= 20  # Scale x, y, z to be within reasonable range
current_vertex[:, 3] = np.abs(current_vertex[:, 3])  # Intensity should be positive

ARGS = {
    'current_vertex': current_vertex,
    'fov_up': 3.0,
    'fov_down': -25.0,
    'proj_H': 64,
    'proj_W': 900,
    'max_range': 50
}
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
map Not yet run
unverified (harvested, no recorded run)
blurryface0814/pcpnet/pcpnet/utils/logger.py
code served (permissive licence) · get_code("779a95e75f5ef68f")
How it ran

no invocation recorded.

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, 8 are also present in other papers whose code we hold (8 with at least one paper dated earlier, 0 the same month, 0 later) and 0 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_filename 1 other paper
1 dated earlier
2110.04076 (2021-10)
Self-supervised Point Cloud Prediction Using 3D Spatio-temporal Convolutional Networks
get_pred_and_gt_point_cloud 1 other paper
1 dated earlier
2110.04076 (2021-10)
Self-supervised Point Cloud Prediction Using 3D Spatio-temporal Convolutional Networks
last_file 1 other paper
1 dated earlier
2110.04076 (2021-10)
Self-supervised Point Cloud Prediction Using 3D Spatio-temporal Convolutional Networks
load_calib 1 other paper
1 dated earlier
2110.04076 (2021-10)
Self-supervised Point Cloud Prediction Using 3D Spatio-temporal Convolutional Networks
load_poses 1 other paper
1 dated earlier
2110.04076 (2021-10)
Self-supervised Point Cloud Prediction Using 3D Spatio-temporal Convolutional Networks
make_path 1 other paper
1 dated earlier
2110.04076 (2021-10)
Self-supervised Point Cloud Prediction Using 3D Spatio-temporal Convolutional Networks
map 1 other paper
1 dated earlier
2207.12691 (2022-07)
CENet: Toward Concise and Efficient LiDAR Semantic Segmentation for Autonomous Driving
range_projection 1 other paper
1 dated earlier
2110.04076 (2021-10)
Self-supervised Point Cloud Prediction Using 3D Spatio-temporal Convolutional Networks

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

The ability to predict future structure features of environments based on past perception information is extremely needed by autonomous vehicles, which helps to make the following decision-making and path planning more reasonable. Recently, point cloud prediction (PCP) is utilized to predict and describe future environmental structures by the point cloud form. In this letter, we propose a novel efficient Transformer-based network to predict the future LiDAR point clouds exploiting the past point cloud sequences. We also design a semantic auxiliary training strategy to make the predicted LiDAR point cloud sequence semantically similar to the ground truth and thus improves the significance of the deployment for more tasks in real-vehicle applications. Our approach is completely self-supervised, which means it does not require any manual labeling and has a solid generalization ability toward different environments. The experimental results show that our method outperforms the state-of-the-art PCP methods on the prediction results and semantic similarity, and has a good real-time performance. Our open-source code and pre-trained models are available at https://github.com/Blurryface0814/PCPNet.

For agents

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

get_harvested_code_for_paper("2304.07773")
get_code_for_paper("2304.07773")
have("2304.07773")

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

curl https://syntology.ai/api/ran/2304.07773.json

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

[![Syntology run record](https://syntology.ai/api/ran/2304.07773.svg)](https://syntology.ai/paper/2304.07773)

Connect an agent — have() is free.