SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 1812.05784 · 2018

PointPillars: Fast Encoders for Object Detection from Point Clouds

arXiv · PDF · Open in the Atlas

Code that ran

13 of 15 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 9 have no contract verdict; 4 failures of our instrument; 2 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/1812.05784.json · badge: /api/ran/1812.05784.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
SmallMunich/nutonomy_pointpillars community (classifier label)
classifier basis not recorded on the link
6 ran · 2 no recorded run
LKLQQ/pointpillars community (classifier label)
classifier basis not recorded on the link
3 ran · 0 no recorded run
airalcorn2/paved2paradise community (classifier label)
classifier basis not recorded on the link
2 ran · 0 no recorded run
zhulf0804/pointpillars reimplementation (classifier label)
classifier basis: classifier_2025
1 ran · 0 no recorded run
isl-org/Open3D-ML extension (classifier label)
classifier basis: classifier_2025
1 ran · 0 no recorded run
FunctionStatusWhere it lives
create_depth_image Ran
ran on a synthesized input
airalcorn2/paved2paradise/paved2paradise.py
code served (permissive licence) · get_code("566cdbac52dacce0")
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 Image. 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
from PIL import Image

torch.manual_seed(0)

# Create a small depth array
# The function expects depths to support:
# - indexing with boolean mask (depths[depths == 0])
# - division by scalar (depths / max_depth)
# - astype conversion
# This works with numpy arrays
depths = np.array([[1.5, 2.0, 0.0, 3.5],
                   [0.0, 2.5, 1.8, 2.2]], dtype=np.float32)

max_depth = 5.0

ARGS = {
    "depths": depths,
    "max_depth": max_depth
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
PIL, numpy, torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Image — a Image output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a Image output the recorded ndim 2 and dtype family int 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_config Ran
ran on a synthesized input
LKLQQ/pointpillars/src/utils.py
code served (permissive licence) · get_code("17815b44aafcf555")
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 yaml
import tempfile
import os

# Create a temporary YAML file with valid configuration
temp_dir = tempfile.mkdtemp()
cfg_path = os.path.join(temp_dir, "config.yaml")

# Write a sample YAML configuration
config_content = """
model:
  name: test_model
  layers: 3
  hidden_dim: 64
training:
  batch_size: 32
  learning_rate: 0.001
  epochs: 10
"""

with open(cfg_path, 'w') as f:
    f.write(config_content)

ARGS = {
    'cfg_path': cfg_path
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
yaml — 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_dataset_keys Ran
ran on a synthesized input
LKLQQ/pointpillars/src/builder/dataset_builder.py
code served (permissive licence) · get_code("4967576c687dfdf8")
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 list, ndim 1, dtype family float; no output shape is stated. The image it ran in was not recorded. If your copy returns the same type and ndim on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch

torch.manual_seed(0)

ARGS = {
    'training': True,
    'generate_bev': True,
}
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 list, ndim 1, dtype family float — a list output has no array shape of its own; none is stated
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_params_for_net Ran
ran on a synthesized input
LKLQQ/pointpillars/src/utils.py
code served (permissive licence) · get_code("451d86db9edf4584")
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)

# Construct a params dict that mimics what get_params_for_net expects
# The function filters for keys starting with 'optimizer.' and 'network.network.'
# and extracts the suffix after removing the prefix

params = {
    'optimizer.learning_rate': 0.001,
    'optimizer.weight_decay': 0.0001,
    'optimizer.momentum': 0.9,
    'network.network.layer1.weight': torch.randn(3, 4),
    'network.network.layer1.bias': torch.randn(3),
    'network.network.layer2.weight': torch.randn(2, 3),
    'other.param': 'ignored',
    'unrelated': 42,
}

ARGS = {
    'params': params
}
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_score_thresholds Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
zhulf0804/pointpillars/evaluate.py
code served (permissive licence) · get_code("a48052245131c33f")
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 list, ndim 1, dtype family float; no output shape is stated. The image it ran in was not recorded. If your copy returns the same type and ndim on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
import numpy as np

torch.manual_seed(0)

# Analyze the function:
# - tp_scores: needs to be a list/array of numeric values (scores)
# - total_num_valid_gt: needs to be an integer (total number of valid ground truth)
# - num_sample_pts: optional parameter with default 41, needs to be an integer

# The function sorts tp_scores in descending order and iterates through them
# It uses len(tp_scores), so tp_scores should support len()
# It performs arithmetic operations on scores and indices

# Create valid arguments:
tp_scores = [0.95, 0.87, 0.76, 0.64, 0.52, 0.41, 0.38, 0.25, 0.12, 0.08]
total_num_valid_gt = 15
num_sample_pts = 41

ARGS = {
    "tp_scores": tp_scores,
    "total_num_valid_gt": total_num_valid_gt,
    "num_sample_pts": num_sample_pts,
}
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 list, ndim 1, dtype family float — a list output has no array shape of its own; none is stated
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
is_line_segment_cross Ran
ran on a synthesized input
SmallMunich/nutonomy_pointpillars/second/core/geometry.py
code served (permissive licence) · get_code("abc7a1803bf35228")
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 (2, 2) as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import numpy as np

# The function is_line_segment_cross takes two arrays of line segments
# lines1 has shape (n1, 2, 2) where each line is defined by two points [start, end]
# lines2 has shape (n2, 2, 2) where each line is defined by two points [start, end]
# Each point has 2 coordinates (x, y)

lines1 = np.array([
    [[0.0, 0.0], [1.0, 1.0]],
    [[2.0, 0.0], [3.0, 1.0]]
], dtype=np.float32)

lines2 = np.array([
    [[0.0, 1.0], [1.0, 0.0]],
    [[2.0, 1.0], [3.0, 0.0]]
], dtype=np.float32)

ARGS = {
    "lines1": lines1,
    "lines2": lines2
}
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 (2, 2) — shape (2, 2) as a later run recorded it
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
is_line_segment_intersection_jit Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
SmallMunich/nutonomy_pointpillars/second/core/geometry.py
code served (permissive licence) · get_code("de9bc527072ac4b0")
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 (2, 3) as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import numpy as np
import numba

# Create sample line segments
# lines1: N=2 line segments, each defined by 2 points with 2D coordinates
lines1 = np.array([
    [[0.0, 0.0], [1.0, 1.0]],  # Line segment 1
    [[0.0, 1.0], [1.0, 0.0]]   # Line segment 2
], dtype=np.float64)

# lines2: M=3 line segments, each defined by 2 points with 2D coordinates
lines2 = np.array([
    [[0.5, 0.0], [0.5, 1.0]],  # Line segment 3
    [[0.0, 0.5], [1.0, 0.5]],  # Line segment 4
    [[2.0, 2.0], [3.0, 3.0]]   # Line segment 5 (non-intersecting)
], dtype=np.float64)

ARGS = {
    "lines1": lines1,
    "lines2": lines2
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numba, 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 (2, 3) — shape (2, 3) as a later run recorded it
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
line_segment_intersection Ran
ran on a synthesized input
SmallMunich/nutonomy_pointpillars/second/core/geometry.py
code served (permissive licence) · get_code("6a7ad14559c54076")
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 bool, ndim 0, dtype family float, 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 numpy as np
import numba

# Create numpy arrays for the line segments
# line1 represents a line segment from point A to point B
# line2 represents a line segment from point C to point D
line1 = np.array([[0.0, 0.0], [2.0, 2.0]], dtype=np.float64)
line2 = np.array([[0.0, 2.0], [2.0, 0.0]], dtype=np.float64)

# Create an output array to store the intersection point
intersection = np.array([0.0, 0.0], dtype=np.float64)

ARGS = {
    "line1": line1,
    "line2": line2,
    "intersection": intersection
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numba, numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type bool, ndim 0, dtype family float, 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
second_box_decode Ran
ran on a synthesized input
SmallMunich/nutonomy_pointpillars/second/core/box_np_ops.py
code served (permissive licence) · get_code("526cf943db7338d3")
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 (2, 7) 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

# Create small test data matching the function signature
# The function expects:
# - box_encodings: [N, 7] or [N, 8] array depending on encode_angle_to_vector
# - anchors: [N, 7] array
# - encode_angle_to_vector: boolean
# - smooth_dim: boolean

N = 2  # batch size

# Create anchors: [N, 7] with x, y, z, w, l, h, r
anchors = np.array([
    [0.0, 0.0, 1.0, 2.0, 3.0, 1.5, 0.5],
    [1.0, 1.0, 2.0, 2.5, 3.5, 2.0, 1.0]
], dtype=np.float32)

# Create box_encodings: [N, 7] for encode_angle_to_vector=False
box_encodings_7 = np.array([
    [0.1, 0.2, 0.3, 0.1, 0.1, 0.1, 0.2],
    [0.2, 0.1, 0.15, 0.05, 0.05, 0.05, 0.1]
], dtype=np.float32)

# Create box_encodings: [N, 8] for encode_angle_to_vector=True
box_encodings_8 = np.array([
    [0.1, 0.2, 0.3, 0.1, 0.1, 0.1, 0.1, 0.2],
    [0.2, 0.1, 0.15, 0.05, 0.05, 0.05, 0.05, 0.1]
], dtype=np.float32)

# Test case 1: encode_angle_to_vector=False, smooth_dim=False
ARGS = {
    "box_encodings": box_encodings_7,
    "anchors": anchors,
    "encode_angle_to_vector": False,
    "smooth_dim": False
}
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 (2, 7) — shape (2, 7) 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
second_box_encode Ran
ran on a synthesized input
SmallMunich/nutonomy_pointpillars/second/core/box_np_ops.py
code served (permissive licence) · get_code("f2f3e2a705ddc55f")
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 (2, 7) 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

# Create small sample data for boxes and anchors
# Both should be [N, 7] tensors: x, y, z, w, l, h, r
np.random.seed(0)

# Create 2 boxes and 2 anchors with 7 dimensions each
boxes = np.random.randn(2, 7).astype(np.float32)
anchors = np.random.randn(2, 7).astype(np.float32)

# Ensure positive dimensions (w, l, h are typically positive)
boxes[:, 3:6] = np.abs(boxes[:, 3:6]) + 0.5
anchors[:, 3:6] = np.abs(anchors[:, 3:6]) + 0.5

ARGS = {
    "boxes": boxes,
    "anchors": anchors,
    "encode_angle_to_vector": False,
    "smooth_dim": False,
}
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 (2, 7) — shape (2, 7) 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
unlevel Ran
ran on a synthesized input
airalcorn2/paved2paradise/paved2paradise.py
code served (permissive licence) · get_code("c61b1e019a047572")
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 numpy as np

torch.manual_seed(0)

class BBox:
    def __init__(self, center, R):
        self.center = center
        self.R = R

# Create small tensors with distinct dimensions for clarity
moved_object_points = torch.randn(2, 3, dtype=torch.float32)  # 2 points, 3D coordinates
bg_t = torch.randn(3, dtype=torch.float32)  # translation vector, 3D
bg_R = torch.randn(3, 3, dtype=torch.float32)  # rotation matrix, 3x3
bbox_center = torch.randn(3, dtype=torch.float32)  # center point, 3D
bbox_R = torch.randn(3, 3, dtype=torch.float32)  # rotation matrix, 3x3
moved_bbox = BBox(center=bbox_center, R=bbox_R)

ARGS = {
    "moved_object_points": moved_object_points,
    "bg_t": bg_t,
    "bg_R": bg_R,
    "moved_bbox": moved_bbox,
}
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 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
unmap Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
SmallMunich/nutonomy_pointpillars/second/core/target_ops.py
code served (permissive licence) · get_code("2902bedb7c18b1a1")
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 1, dtype family int, shape (5,) as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import numpy as np

# Test case 1: 1D array with subset indices
data_1d = np.array([10, 20, 30], dtype=np.int32)
count_1d = 5
inds_1d = np.array([1, 3, 4], dtype=np.int64)
fill_1d = 0

# Test case 2: 2D array with subset indices
data_2d = np.array([[1, 2], [3, 4], [5, 6]], dtype=np.float32)
count_2d = 4
inds_2d = np.array([0, 2, 3], dtype=np.int64)
fill_2d = 0

# Use case 1 for ARGS
ARGS = {
    "data": data_1d,
    "count": count_1d,
    "inds": inds_1d,
    "fill": fill_1d
}
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 1, dtype family int, shape (5,) — shape (5,) as a later run recorded it
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
unpack Ran
ran; our contract draft was wrong, not the code
isl-org/Open3D-ML/ml3d/tf/models/point_pillars.py
pointer only (licence: NOASSERTION) · get_code("1354276502be4aa8")
How it ran

The body of this sample is withheld: GitHub could not classify the licence of this paper's repository, isl-org/Open3D-ML (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. A run is recorded.

create_target_np Not yet run
unverified (harvested, no recorded run)
SmallMunich/nutonomy_pointpillars/second/core/target_ops.py
code served (permissive licence) · get_code("d2b64ea77c2d6d9f")
How it ran

no invocation recorded.

riou_cc Not yet run
unverified (harvested, no recorded run)
SmallMunich/nutonomy_pointpillars/second/core/box_np_ops.py
code served (permissive licence) · get_code("2b3e25e607e98a0f")
How it ran

no invocation recorded.

Also present in other papers' code

Of 15 samples lifted from this paper's linked repositories, 15 carry a function identity and 0 do not. Of the 15 distinct functions with an identity, 7 are also present in other papers whose code we hold (0 with at least one paper dated earlier, 0 the same month, 7 later) and 8 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
unmap 4 other papers
4 later
1908.09492 (2019-08)
Class-balanced Grouping and Sampling for Point Cloud 3D Object Detection
1912.04986 (2019-12)
What You See is What You Get: Exploiting Visibility for 3D Object Detection
2009.00784 (2020-09)
CLOCs: Camera-LiDAR Object Candidates Fusion for 3D Object Detection
and 1 more not named here
is_line_segment_intersection_jit 3 other papers
3 later
1908.09492 (2019-08)
Class-balanced Grouping and Sampling for Point Cloud 3D Object Detection
1912.04986 (2019-12)
What You See is What You Get: Exploiting Visibility for 3D Object Detection
2009.00784 (2020-09)
CLOCs: Camera-LiDAR Object Candidates Fusion for 3D Object Detection
line_segment_intersection 3 other papers
3 later
1908.09492 (2019-08)
Class-balanced Grouping and Sampling for Point Cloud 3D Object Detection
1912.04986 (2019-12)
What You See is What You Get: Exploiting Visibility for 3D Object Detection
2009.00784 (2020-09)
CLOCs: Camera-LiDAR Object Candidates Fusion for 3D Object Detection
is_line_segment_cross 2 other papers
2 later
1912.04986 (2019-12)
What You See is What You Get: Exploiting Visibility for 3D Object Detection
2009.00784 (2020-09)
CLOCs: Camera-LiDAR Object Candidates Fusion for 3D Object Detection
create_target_np 1 other paper
1 later
2009.00784 (2020-09)
CLOCs: Camera-LiDAR Object Candidates Fusion for 3D Object Detection
second_box_decode 1 other paper
1 later
2009.00784 (2020-09)
CLOCs: Camera-LiDAR Object Candidates Fusion for 3D Object Detection
second_box_encode 1 other paper
1 later
2009.00784 (2020-09)
CLOCs: Camera-LiDAR Object Candidates Fusion for 3D Object Detection

Presence, not direction: an identical function body in two papers' repositories does not say who wrote it, which came from which, or that either paper's method rests on it. Matching is exact after normalisation, so a renamed or edited function is not matched. Dates are when papers appeared, not when code was written. Only .py files the harvester lifted are covered; notebooks are not read. A new-style arXiv id YYMM.NNNN dates a paper to the month 20YY-MM; otherwise Paper.year dates it to the year; otherwise it is undated. Dates are when the paper appeared, not when the code was written. Read 2026-11-22.

Repositories linked to this paper

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

Abstract

Object detection in point clouds is an important aspect of many robotics applications such as autonomous driving. In this paper we consider the problem of encoding a point cloud into a format appropriate for a downstream detection pipeline. Recent literature suggests two types of encoders; fixed encoders tend to be fast but sacrifice accuracy, while encoders that are learned from data are more accurate, but slower. In this work we propose PointPillars, a novel encoder which utilizes PointNets to learn a representation of point clouds organized in vertical columns (pillars). While the encoded features can be used with any standard 2D convolutional detection architecture, we further propose a lean downstream network. Extensive experimentation shows that PointPillars outperforms previous encoders with respect to both speed and accuracy by a large margin. Despite only using lidar, our full detection pipeline significantly outperforms the state of the art, even among fusion methods, with respect to both the 3D and bird's eye view KITTI benchmarks. This detection performance is achieved while running at 62 Hz: a 2 - 4 fold runtime improvement. A faster version of our method matches the state of the art at 105 Hz. These benchmarks suggest that PointPillars is an appropriate encoding for object detection in point clouds.

For agents

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

get_harvested_code_for_paper("1812.05784")
get_code_for_paper("1812.05784")
have("1812.05784")

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

curl https://syntology.ai/api/ran/1812.05784.json

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

[![Syntology run record](https://syntology.ai/api/ran/1812.05784.svg)](https://syntology.ai/paper/1812.05784)

Connect an agent — have() is free.