SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2308.04409 · 2023

V-DETR: DETR with Vertex Relative Position Encoding for 3D Object Detection

arXiv · PDF · Open in the Atlas

Code that ran

10 of 12 samples (functions and classes) lifted from this paper's linked repositories executed on a synthesized input in Syntology's jail (0 of them constructed an object rather than computing a result; 2 honoured the contract we drafted, 1 violated it and 0 have no contract verdict; 7 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/2308.04409.json · badge: /api/ran/2308.04409.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
yichaoshen-ms/v-detr canonical (classifier label)
classifier basis: census
10 ran · 2 no recorded run
FunctionStatusWhere it lives
adjust_learning_rate Ran
ran, honoured the contract we drafted
yichaoshen-ms/v-detr/engine.py
code served (permissive licence) · get_code("2908efab7e71b5a8")
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 float, 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 torch
import torch.nn as nn
from torch.optim import SGD
import math

torch.manual_seed(0)

class Args:
    def __init__(self):
        self.warm_lr_epochs = 5
        self.max_epoch = 100
        self.warm_lr = 0.001
        self.base_lr = 0.1
        self.final_lr = 0.0001
        self.lr_scheduler = 'cosine'
        self.step_epoch = '30_60'

args_obj = Args()

class SimpleModel(nn.Module):
    def __init__(self):
        super().__init__()
        self.linear = nn.Linear(4, 3)
    
    def forward(self, x):
        return self.linear(x)

model = SimpleModel()
optimizer = SGD(model.parameters(), lr=0.1)

curr_epoch = 0.5

ARGS = {
    'args': args_obj,
    'optimizer': optimizer,
    'curr_epoch': curr_epoch
}
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 float, 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
build_optimizer Ran
ran; our contract draft was wrong, not the code
yichaoshen-ms/v-detr/optimizer.py
code served (permissive licence) · get_code("da4f8d084fea1b3e")
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 AdamW. 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 SimpleModel(nn.Module):
    def __init__(self):
        super().__init__()
        self.linear1 = nn.Linear(4, 3)
        self.linear2 = nn.Linear(3, 2)
        self.bias_param = nn.Parameter(torch.randn(5))
    
    def forward(self, x):
        return self.linear2(self.linear1(x))

model = SimpleModel()

class Args:
    def __init__(self):
        self.filter_biases_wd = True
        self.weight_decay = 0.01
        self.base_lr = 0.001

args = Args()

ARGS = {
    'args': args,
    'model': model
}
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 AdamW — a AdamW output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a AdamW 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
compute_learning_rate Ran
ran, honoured the contract we drafted
yichaoshen-ms/v-detr/engine.py
code served (permissive licence) · get_code("38f2153fcc1523d8")
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 float, 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 math

class Args:
    def __init__(self):
        self.warm_lr_epochs = 2
        self.max_epoch = 10
        self.warm_lr = 0.0001
        self.base_lr = 0.1
        self.final_lr = 0.001
        self.lr_scheduler = 'cosine'
        self.step_epoch = '3_7'

ARGS = {
    'args': Args(),
    'curr_epoch_normalized': 0.5
}
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 float, 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
convert_corners_camera2lidar Ran
ran, violated the contract we drafted
yichaoshen-ms/v-detr/models/vdetr_transformer.py
code served (permissive licence) · get_code("c442c7a276779ddf")
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 3, dtype family float, shape (2, 4, 3) as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
import numpy as np

torch.manual_seed(0)

# The function takes corners_camera and performs in-place operations:
# 1. Negates the second dimension (index 1)
# 2. Permutes dimensions 0, 1, 2 to order 0, 2, 1
# corners_camera should be a tensor with shape (..., 3) where the last dimension
# has at least 3 elements (x, y, z coordinates)

# Create a small tensor representing camera corners
# Shape: (2, 4, 3) - batch of 2, 4 corners per object, 3 coordinates each
corners_camera = torch.randn(2, 4, 3)

ARGS = {
    "corners_camera": corners_camera
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Tensor, ndim 3, dtype family float, shape (2, 4, 3) — shape (2, 4, 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
get_clones Ran
ran; our contract draft was wrong, not the code
yichaoshen-ms/v-detr/models/helpers.py
pointer only (licence: MIT); the stored copy get_code serves, from arshadshk/SAINT-pytorch, is pointer only (licence: NONE) · get_code("891b8ebab395921f")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (MIT), but get_code serves the stored copy, first harvested from arshadshk/SAINT-pytorch, 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.

inverse_sigmoid Ran
ran; our contract draft was wrong, not the code
yichaoshen-ms/v-detr/models/vdetr_transformer.py
pointer only (licence: MIT); the stored copy get_code serves, from woodfrog/poly-diffuse, is pointer only (licence: GPL-3.0) · get_code("83c03ceb1307c6c4")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (MIT), but get_code serves the stored copy, first harvested from woodfrog/poly-diffuse, and GPL-3.0, as GitHub detects it on that repository, is not in the permissive set Syntology serves inline. 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.

read_plymesh Ran
ran; our contract draft was wrong, not the code
yichaoshen-ms/v-detr/datasets/scannet.py
code served (permissive licence) · get_code("484cab6df3da608c")
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
import pandas as pd
import plyfile
import tempfile
import os

# Create a temporary PLY file for testing
def create_test_ply_file():
    """Create a minimal valid PLY file."""
    temp_dir = tempfile.gettempdir()
    filepath = os.path.join(temp_dir, "test_mesh.ply")
    
    # Create a simple triangle mesh
    vertices = np.array([(0, 0, 0), (1, 0, 0), (0, 1, 0)], 
                        dtype=[('x', 'f4'), ('y', 'f4'), ('z', 'f4')])
    faces = np.array([([0, 1, 2],)], 
                     dtype=[('vertex_indices', 'i4', (3,))])
    
    # Create PlyData object
    vertex_el = plyfile.PlyElement.describe(vertices, 'vertex')
    face_el = plyfile.PlyElement.describe(faces, 'face')
    ply_data = plyfile.PlyData([vertex_el, face_el])
    
    # Write to file
    ply_data.write(filepath)
    
    return filepath

test_filepath = create_test_ply_file()

ARGS = {
    'filepath': test_filepath
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, pandas, plyfile — 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
read_segmentation Ran
ran; our contract draft was wrong, not the code
yichaoshen-ms/v-detr/datasets/scannet.py
code served (permissive licence) · get_code("395ce1c22ca13d08")
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 os
import json
import tempfile

# Create a temporary JSON file with valid segmentation data
temp_dir = tempfile.mkdtemp()
temp_filename = os.path.join(temp_dir, "segmentation.json")

# Create sample segmentation data
seg_data = {
    "segIndices": [0, 1, 0, 2, 1, 2, 0, 1]
}

with open(temp_filename, 'w') as f:
    json.dump(seg_data, f)

ARGS = {
    "filename": temp_filename
}
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 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
roty_batch_tensor Ran
ran; our contract draft was wrong, not the code
yichaoshen-ms/v-detr/models/vdetr_transformer.py
code served (permissive licence) · get_code("f99bc72a07da96e0")
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 4, dtype family float, shape (2, 4, 3, 3) as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch

torch.manual_seed(0)

# The function roty_batch_tensor takes a tensor t of angles and produces
# rotation matrices. It supports batched inputs of arbitrary shape.
# The output will be shape (*t.shape, 3, 3)

# Create a small tensor of angles with batch dimensions
t = torch.randn(2, 4)  # batch size 2, 4 angles per batch

ARGS = {
    "t": t
}
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 4, dtype family float, shape (2, 4, 3, 3) — shape (2, 4, 3, 3) as a later run recorded it
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
sigmoid_focal_loss Ran
ran; our contract draft was wrong, not the code
yichaoshen-ms/v-detr/criterion.py
code served (permissive licence) · get_code("5c0711aada67957e")
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) and returned Tensor, 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 torch
import torch.nn.functional as F

torch.manual_seed(0)

# Create small tensors for inputs and targets
# inputs: predictions (logits) - shape (2, 4) for batch_size=2, num_predictions=4
inputs = torch.randn(2, 4, requires_grad=True)

# targets: binary labels - shape (2, 4) matching inputs
targets = torch.randint(0, 2, (2, 4), dtype=torch.float32)

# num_boxes: number of boxes for normalization
num_boxes = 2

# alpha and gamma are already float parameters with defaults
alpha = 0.25
gamma = 2.0

ARGS = {
    "inputs": inputs,
    "targets": targets,
    "num_boxes": num_boxes,
    "alpha": alpha,
    "gamma": gamma,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1
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 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
convert_unnorm2norm Not yet run
unverified (harvested, no recorded run)
yichaoshen-ms/v-detr/models/model_vdetr.py
code served (permissive licence) · get_code("47cc74bdb126783e")
How it ran

no invocation recorded.

get_release_scans Not yet run
unverified (harvested, no recorded run)
yichaoshen-ms/v-detr/scannet/download-scannet.py
code served (permissive licence) · get_code("1d1b94d9af6bae68")
How it ran

no invocation recorded.

Also present in other papers' code

Of 12 samples lifted from this paper's linked repositories, 12 carry a function identity and 0 do not. Of the 12 distinct functions with an identity, 7 are also present in other papers whose code we hold (4 with at least one paper dated earlier, 3 the same month, 6 later) and 5 are only here among the 52,977 papers whose code Syntology holds (as of 2026-09-26).

Only here means: among the 52,977 papers whose code Syntology holds (51,887 with lifted samples, 1,090 through their assemblies only), as of 2026-09-26, no other paper's lifted samples or current assembly bricks contain an identical function body after normalisation.

FunctionAlso present inEarliest-datedOthers named
sigmoid_focal_loss 56 other papers
31 dated earlier, 2 the same month, 20 later, 3 same year, order not determinable · 1 only as an assembly brick
2005.12872 (2020-05)
End-to-End Object Detection with Transformers
2010.04159 (2020-10)
Deformable DETR: Deformable Transformers for End-to-End Object Detection
2103.11731 (2021-03)
Meta-DETR: Image-Level Few-Shot Object Detection with Inter-Class Correlation Exploitation
and 53 more not named here
inverse_sigmoid 51 other papers
28 dated earlier, 3 the same month, 20 later · 23 only as an assembly brick
2005.12872 (2020-05)
End-to-End Object Detection with Transformers
2010.04159 (2020-10)
Deformable DETR: Deformable Transformers for End-to-End Object Detection
2103.11161 (2021-03)
MonteFloor: Extending MCTS for Reconstructing Accurate Large-Scale Floor Plans
and 48 more not named here
get_clones 40 other papers
18 dated earlier, 1 the same month, 20 later, 1 same year, order not determinable · 8 only as an assembly brick
1905.12794 (2019-05)
Fashion IQ: A New Dataset Towards Retrieving Images by Natural Language Feedback
2002.07033 (2020-02)
Towards an Appropriate Query, Key, and Value Computation for Knowledge Tracing
2010.12042 (2020-10)
SAINT+: Integrating Temporal Features for EdNet Correctness Prediction
and 37 more not named here
adjust_learning_rate 4 other papers
4 later
2311.18651 (2023-11)
LL3DA: Visual Interactive Instruction Tuning for Omni-3D Understanding, Reasoning, and Planning
2406.00830 (2024-06)
Collaborative Novel Object Discovery and Box-Guided Cross-Modal Alignment for Open-Vocabulary 3D Object Detection
2407.08931 (2024-07)
Global-Local Collaborative Inference with LLM for Lidar-Based Open-Vocabulary Detection
and 1 more not named here
build_optimizer 1 other paper
1 later
2407.08931 (2024-07)
Global-Local Collaborative Inference with LLM for Lidar-Based Open-Vocabulary Detection
convert_corners_camera2lidar 1 other paper
1 later
2503.14493 (2025-03)
State Space Model Meets Transformer: A New Paradigm for 3D Object Detection
read_plymesh 1 other paper
1 dated earlier
2306.03908 (2023-06)
SAM3D: Segment Anything in 3D Scenes

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

We introduce a highly performant 3D object detector for point clouds using the DETR framework. The prior attempts all end up with suboptimal results because they fail to learn accurate inductive biases from the limited scale of training data. In particular, the queries often attend to points that are far away from the target objects, violating the locality principle in object detection. To address the limitation, we introduce a novel 3D Vertex Relative Position Encoding (3DV-RPE) method which computes position encoding for each point based on its relative position to the 3D boxes predicted by the queries in each decoder layer, thus providing clear information to guide the model to focus on points near the objects, in accordance with the principle of locality. In addition, we systematically improve the pipeline from various aspects such as data normalization based on our understanding of the task. We show exceptional results on the challenging ScanNetV2 benchmark, achieving significant improvements over the previous 3DETR in $\rm{AP}_{25}$/$\rm{AP}_{50}$ from 65.0\%/47.0\% to 77.8\%/66.0\%, respectively. In addition, our method sets a new record on ScanNetV2 and SUN RGB-D datasets.Code will be released at http://github.com/yichaoshen-MS/V-DETR.

For agents

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

get_harvested_code_for_paper("2308.04409")
get_code_for_paper("2308.04409")
have("2308.04409")

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

curl https://syntology.ai/api/ran/2308.04409.json

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

[![Syntology run record](https://syntology.ai/api/ran/2308.04409.svg)](https://syntology.ai/paper/2308.04409)

Connect an agent — have() is free.