SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2112.02205 · 2021

Behind the Curtain: Learning Occluded Shapes for 3D Object Detection

arXiv · PDF · Open in the Atlas

Code that ran

5 of 7 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 4 have no contract verdict; 1 failure 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/2112.02205.json · badge: /api/ran/2112.02205.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
xharlie/btcdet canonical (classifier label)
classifier basis not recorded on the link
5 ran · 2 no recorded run
FunctionStatusWhere it lives
cfg_from_yaml_file Ran
ran on a synthesized input
xharlie/btcdet/btcdet/config.py
code served (permissive licence) · get_code("3778ab300cc46205")
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 EasyDict. 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 tempfile
import os
import yaml
from easydict import EasyDict

# Create a temporary YAML file for testing
temp_dir = tempfile.mkdtemp()
base_config_path = os.path.join(temp_dir, 'base_config.yaml')
cfg_file_path = os.path.join(temp_dir, 'config.yaml')

# Write base config
base_config_data = {
    'model': {
        'type': 'resnet',
        'depth': 50
    },
    'learning_rate': 0.001,
    'batch_size': 32
}
with open(base_config_path, 'w') as f:
    yaml.dump(base_config_data, f)

# Write main config that references base
main_config_data = {
    '_BASE_CONFIG_': base_config_path,
    'model': {
        'depth': 101
    },
    'learning_rate': 0.0001,
    'epochs': 100
}
with open(cfg_file_path, 'w') as f:
    yaml.dump(main_config_data, f)

# Create initial config dict
initial_config = EasyDict()

ARGS = {
    'cfg_file': cfg_file_path,
    'config': initial_config
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
easydict, yaml — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type EasyDict — a EasyDict output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a EasyDict 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
extract_allpnts Ran
ran on a synthesized input
xharlie/btcdet/btcdet/datasets/finddiff.py
code served (permissive licence) · get_code("ccbe0fc1350b4cc2")
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 pickle
from pathlib import Path
import tempfile
import os

# Create temporary directory and pickle files for the function to load
temp_dir = tempfile.mkdtemp()

# Create mock database info pickles for 'train' and 'val' splits
for split in ['train', 'val']:
    db_info_data = {
        'Car': [
            {
                'name': 'Car',
                'box3d_lidar': [1.0, 2.0, 3.0, 4.0, 5.0, 6.0, 7.0]
            },
            {
                'name': 'Car',
                'box3d_lidar': [1.5, 2.5, 3.5, 4.5, 5.5, 6.5, 7.5]
            },
            {
                'name': 'Pedestrian',
                'box3d_lidar': [0.5, 1.5, 2.5, 3.5, 4.5, 5.5, 6.5]
            }
        ]
    }
    
    save_path = Path(temp_dir) / ('kitti_dbinfos_%s.pkl' % split)
    with open(save_path, 'wb') as f:
        pickle.dump(db_info_data, f)

ARGS = {
    'root_path': temp_dir,
    'splits': ['train', 'val']
}
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 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
extract_template Ran
ran on a synthesized input
xharlie/btcdet/btcdet/datasets/template.py
code served (permissive licence) · get_code("9e6494b06ede8d3a")
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 pickle
from pathlib import Path
import numpy as np
import tempfile
import os

# Create temporary directory and mock pickle files
temp_dir = tempfile.mkdtemp()

# Create mock db_infos for 'train' split
train_db_infos = [
    {
        'name': 'Car',
        'box3d_lidar': np.array([1.0, 2.0, 3.0, 4.5, 3.2, 1.8]),
    },
    {
        'name': 'Car',
        'box3d_lidar': np.array([2.0, 3.0, 4.0, 4.6, 3.1, 1.9]),
    },
    {
        'name': 'Pedestrian',
        'box3d_lidar': np.array([0.5, 1.0, 2.0, 0.8, 0.7, 1.7]),
    },
]

train_pickle_data = {'Car': train_db_infos}
train_path = Path(temp_dir) / 'kitti_dbinfos_train.pkl'
with open(train_path, 'wb') as f:
    pickle.dump(train_pickle_data, f)

# Create mock db_infos for 'val' split
val_db_infos = [
    {
        'name': 'Car',
        'box3d_lidar': np.array([1.5, 2.5, 3.5, 4.4, 3.3, 1.7]),
    },
    {
        'name': 'Car',
        'box3d_lidar': np.array([2.5, 3.5, 4.5, 4.7, 3.0, 2.0]),
    },
]

val_pickle_data = {'Car': val_db_infos}
val_path = Path(temp_dir) / 'kitti_dbinfos_val.pkl'
with open(val_path, 'wb') as f:
    pickle.dump(val_pickle_data, f)

ARGS = {
    'root_path': temp_dir,
    'splits': ['train', 'val'],
}
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
get_normalized_cloud Ran
ran on a synthesized input
xharlie/btcdet/btcdet/datasets/findbestfit.py
code served (permissive licence) · get_code("26d81d4bef629364")
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 (10, 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 binary file with point cloud data
# The function expects to read from a file using np.fromfile
temp_dir = tempfile.mkdtemp()
temp_file = os.path.join(temp_dir, "points.bin")

# Create sample point cloud data: N points with 4 values each (x, y, z, intensity)
# 10 points, 4 channels (x, y, z, intensity)
sample_points = np.random.randn(10, 4).astype(np.float32)
sample_points.tofile(temp_file)

# Create gt_box: [x, y, z, length, width, height, rotation_angle, ...]
# The function uses: gt_box[5] for height, gt_box[6] for rotation angle
gt_box = np.array([0.0, 0.0, 0.0, 3.0, 2.0, 1.5, 0.5, 0.0], dtype=np.float32)

ARGS = {
    "obj_pnt_fpath": temp_file,
    "gt_box": gt_box,
    "bottom": 0.2
}
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 (10, 4) — shape (10, 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
merge_new_config Ran
ran; our contract draft was wrong, not the code
xharlie/btcdet/btcdet/config.py
pointer only (licence: Apache-2.0); the stored copy get_code serves, from mickeykang16/Ev-Stereo3D, is pointer only (licence: NONE) · get_code("392c0cf3a1b07b12")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (Apache-2.0), but get_code serves the stored copy, first harvested from mickeykang16/Ev-Stereo3D, 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.

clustering Not yet run
unverified (harvested, no recorded run)
xharlie/btcdet/btcdet/datasets/findbestfit.py
code served (permissive licence) · get_code("0b15dc5e864c4718")
How it ran

no invocation recorded.

get_corner_loss_lidar Not yet run
unverified (harvested, no recorded run)
xharlie/btcdet/btcdet/utils/loss_utils.py
code served (permissive licence) · get_code("b7ee7b7b0b6ffed4")
How it ran

no invocation recorded.

Also present in other papers' code

Of 7 samples lifted from this paper's linked repositories, 7 carry a function identity and 0 do not. Of the 7 distinct functions with an identity, 2 are also present in other papers whose code we hold (2 with at least one paper dated earlier, 0 the same month, 2 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_corner_loss_lidar 49 other papers
7 dated earlier, 42 later
1907.03670 (2019-07)
From Points to Parts: 3D Object Detection from Point Cloud with Part-aware and Part-aggregation Network
1912.13192 (2019-12)
PV-RCNN: Point-Voxel Feature Set Abstraction for 3D Object Detection
2005.10821 (2020-05)
Hierarchical Multi-Scale Attention for Semantic Segmentation
and 46 more not named here
merge_new_config 15 other papers
5 dated earlier, 10 later
1912.13192 (2019-12)
PV-RCNN: Point-Voxel Feature Set Abstraction for 3D Object Detection
2005.10821 (2020-05)
Hierarchical Multi-Scale Attention for Semantic Segmentation
2006.11275 (2020-06)
Center-based 3D Object Detection and Tracking
and 12 more not named here

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

Advances in LiDAR sensors provide rich 3D data that supports 3D scene understanding. However, due to occlusion and signal miss, LiDAR point clouds are in practice 2.5D as they cover only partial underlying shapes, which poses a fundamental challenge to 3D perception. To tackle the challenge, we present a novel LiDAR-based 3D object detection model, dubbed Behind the Curtain Detector (BtcDet), which learns the object shape priors and estimates the complete object shapes that are partially occluded (curtained) in point clouds. BtcDet first identifies the regions that are affected by occlusion and signal miss. In these regions, our model predicts the probability of occupancy that indicates if a region contains object shapes. Integrated with this probability map, BtcDet can generate high-quality 3D proposals. Finally, the probability of occupancy is also integrated into a proposal refinement module to generate the final bounding boxes. Extensive experiments on the KITTI Dataset and the Waymo Open Dataset demonstrate the effectiveness of BtcDet. Particularly, for the 3D detection of both cars and cyclists on the KITTI benchmark, BtcDet surpasses all of the published state-of-the-art methods by remarkable margins. Code is released (https://github.com/Xharlie/BtcDet}{https://github.com/Xharlie/BtcDet).

For agents

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

get_harvested_code_for_paper("2112.02205")
get_code_for_paper("2112.02205")
have("2112.02205")

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

curl https://syntology.ai/api/ran/2112.02205.json

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

[![Syntology run record](https://syntology.ai/api/ran/2112.02205.svg)](https://syntology.ai/paper/2112.02205)

Connect an agent — have() is free.