SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2401.09939 · 2024

ICGNet: A Unified Approach for Instance-Centric Grasping

arXiv · PDF · Open in the Atlas

Code that ran

8 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 0 have no contract verdict; 8 failures of our instrument; 0 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/2401.09939.json · badge: /api/ran/2401.09939.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
renezurbruegg/icg_benchmark canonical (classifier label)
classifier basis: census
8 ran · 0 no recorded run
FunctionStatusWhere it lives
angle_axis_to_rotation_matrix Ran
ran; our contract draft was wrong, not the code
renezurbruegg/icg_benchmark/icg_benchmark/models/edge_grasp/transform.py
code served (permissive licence) · get_code("982a56a2dba92757")
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, 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 expects angle_axis: Tensor of shape (N, 3)
# where N is batch size and 3 is the axis-angle representation
angle_axis = torch.randn(2, 3)

ARGS = {
    "angle_axis": angle_axis,
}
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 3, dtype family float, shape (2, 3, 3) — shape (2, 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
conv Ran
ran; our contract draft was wrong, not the code
renezurbruegg/icg_benchmark/icg_benchmark/models/giga/networks.py
code served (permissive licence) · get_code("70c2537c5427c0e7")
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 Conv3d. 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)

ARGS = {
    "in_channels": 3,
    "out_channels": 16,
    "kernel_size": 3,
}
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 Conv3d — a Conv3d output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a Conv3d 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_geometry_mask Ran
ran; our contract draft was wrong, not the code
renezurbruegg/icg_benchmark/icg_benchmark/models/edge_grasp/utils.py
code served (permissive licence) · get_code("e724c27f11b875d8")
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, ndim 2, 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)

# Create tensor arguments with distinct sizes for clarity
batch_size = 2
feature_dim = 3

# normals_dot: result of dot product, needs squeeze(dim=-1), shape (batch, 1)
normals_dot = torch.randn(batch_size, 1)

# dot_product_2: similar to normals_dot, shape (batch, 1)
dot_product_2 = torch.randn(batch_size, 1)

# relative_pos: position vectors, used in dot products with normals, shape (batch, 3)
relative_pos = torch.randn(batch_size, feature_dim)

# des_normals: desired normals, shape (batch, 3)
des_normals = torch.randn(batch_size, feature_dim)

# sample_normal: sample normal vector, shape (batch, 3)
sample_normal = torch.randn(batch_size, feature_dim)

# sample_pos: sample position, shape (batch, 3)
sample_pos = torch.randn(batch_size, feature_dim)

# pos: point cloud positions for distance computation, shape (num_points, 3)
pos = torch.randn(5, feature_dim)

# strict: boolean flag
strict = False

ARGS = {
    "normals_dot": normals_dot,
    "dot_product_2": dot_product_2,
    "relative_pos": relative_pos,
    "des_normals": des_normals,
    "sample_normal": sample_normal,
    "sample_pos": sample_pos,
    "pos": pos,
    "strict": strict,
}
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, ndim 2, dtype family float — a tuple 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_geometry_mask2 Ran
ran; our contract draft was wrong, not the code
renezurbruegg/icg_benchmark/icg_benchmark/models/edge_grasp/utils.py
code served (permissive licence) · get_code("33d9c1f3879258c8")
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, ndim 2, 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)

# Create small tensors with distinct dimensions for clarity
batch_size = 2
spatial_dim = 3

# normals_dot: result of dot product, needs squeezing on dim=-1
# Shape: (batch_size, 1) before squeeze
normals_dot = torch.randn(batch_size, 1)

# dot_product_2: another dot product result, needs squeezing on dim=-1
# Shape: (batch_size, 1) before squeeze
dot_product_2 = torch.randn(batch_size, 1)

# relative_pos: position differences, used in dot products with des_normals and sample_normal
# Shape: (batch_size, spatial_dim)
relative_pos = torch.randn(batch_size, spatial_dim)

# des_normals: normal vectors, used in dot products and scaling
# Shape: (batch_size, spatial_dim)
des_normals = torch.randn(batch_size, spatial_dim)

# sample_normal: normal vectors at sample point
# Shape: (batch_size, spatial_dim)
sample_normal = torch.randn(batch_size, spatial_dim)

# sample_pos: positions of samples
# Shape: (batch_size, spatial_dim)
sample_pos = torch.randn(batch_size, spatial_dim)

# pos: point cloud positions for collision checking with cdist
# Shape: (num_points, spatial_dim) - different from batch size to show distinction
pos = torch.randn(4, spatial_dim)

ARGS = {
    "normals_dot": normals_dot,
    "dot_product_2": dot_product_2,
    "relative_pos": relative_pos,
    "des_normals": des_normals,
    "sample_normal": sample_normal,
    "sample_pos": sample_pos,
    "pos": pos,
    "strict": False,
}
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, ndim 2, dtype family float — a tuple 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
predict_mesh Ran
ran; our contract draft was wrong, not the code
renezurbruegg/icg_benchmark/icg_benchmark/reconstruction/eval_reconstruction.py
code served (permissive licence) · get_code("cba2aca018134c43")
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)

class MockGenerator(nn.Module):
    def __init__(self):
        super().__init__()
        self.linear = nn.Linear(4, 8)
    
    def generate_mesh(self, input_dict):
        # input_dict has keys: 'inputs', 'normals', 'colors'
        # Return a mock mesh and something else
        pc_input = input_dict['inputs']
        normals = input_dict['normals']
        colors = input_dict['colors']
        
        # Process inputs minimally
        batch_size = pc_input.shape[0]
        
        # Return mock mesh object and auxiliary output
        mock_mesh = {'vertices': torch.randn(batch_size, 3, 4), 'faces': torch.randint(0, 10, (batch_size, 5, 3))}
        auxiliary = None
        
        return mock_mesh, auxiliary

generator = MockGenerator()

# Create small point cloud inputs
batch_size = 2
num_points = 4
point_dim = 3

pc_input = torch.randn(batch_size, num_points, point_dim)
pc_normals = torch.randn(batch_size, num_points, point_dim)
pc_colors = torch.randn(batch_size, num_points, 3)

ARGS = {
    'generator': generator,
    'pc_input': pc_input,
    'pc_normals': pc_normals,
    'pc_colors': pc_colors
}
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
read_data Ran
ran; our contract draft was wrong, not the code
renezurbruegg/icg_benchmark/icg_benchmark/models/edge_grasp/dataset_processor.py
code served (permissive licence) · get_code("69f2d544253cc632")
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 tempfile
import os

# Create a temporary .npz file with the required data
temp_dir = tempfile.mkdtemp()
temp_file = os.path.join(temp_dir, "test_data.npz")

# Create sample data with distinct sizes for different dimensions
pos = np.random.rand(10, 3).astype(np.float32)  # 10 points, 3D coordinates
normals = np.random.rand(10, 3).astype(np.float32)  # 10 normals, 3D
sample = np.random.rand(8, 3).astype(np.float32)  # 8 samples, 3D
radius_p_index = np.array([0, 1, 2, 3, 4], dtype=np.int64)  # indices
radius_p_batch = np.array([0, 0, 1, 1, 1], dtype=np.int64)  # batch assignments
edges = np.array([[0, 1], [1, 2], [2, 3]], dtype=np.int64)  # 3 edges
approaches = np.random.rand(6, 3).astype(np.float32)  # 6 approach vectors, 3D
edge_sample_index = np.array([0, 1, 2, 1], dtype=np.int64)  # 4 mappings
grasp_label = np.array([1, 0, 1, 0, 1], dtype=np.int64)  # 5 labels
depth_projection = np.random.rand(7, 2).astype(np.float32)  # 7 points, 2D projection

# Save as .npz file
np.savez(
    temp_file,
    pos=pos,
    normals=normals,
    sample=sample,
    radius_p_index=radius_p_index,
    radius_p_batch=radius_p_batch,
    edges=edges,
    approachs=approaches,  # Note: function expects 'approachs' not 'approaches'
    edge_sample_index=edge_sample_index,
    grasp_label=grasp_label,
    depth_projection=depth_projection
)

ARGS = {
    "path": temp_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 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
rotation_matrix_to_quaternion Ran
ran; our contract draft was wrong, not the code
renezurbruegg/icg_benchmark/icg_benchmark/models/edge_grasp/transform.py
code served (permissive licence) · get_code("1b4a7734f876b981")
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

torch.manual_seed(0)

# Create a valid rotation matrix tensor with shape (N, 3, 4)
# N=2 (batch size), 3x4 is the required shape
# The 3x3 part should be a valid rotation matrix
rotation_matrix = torch.tensor([
    [[1.0, 0.0, 0.0, 0.0],
     [0.0, 1.0, 0.0, 0.0],
     [0.0, 0.0, 1.0, 0.0]],
    [[0.0, -1.0, 0.0, 0.0],
     [1.0, 0.0, 0.0, 0.0],
     [0.0, 0.0, 1.0, 0.0]]
], dtype=torch.float32)

ARGS = {
    'rotation_matrix': rotation_matrix,
    'eps': 1e-06
}
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
rotation_matrix_to_quaternion_new Ran
ran; our contract draft was wrong, not the code
renezurbruegg/icg_benchmark/icg_benchmark/models/edge_grasp/transform.py
code served (permissive licence) · get_code("d842db640cc94124")
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

torch.manual_seed(0)

# Create a valid rotation matrix tensor of shape (N, 3, 3)
# N=2 for batch size, 3x3 for rotation matrix
rotation_matrix = torch.tensor([
    [[0.9, -0.436, 0.0],
     [0.436, 0.9, 0.0],
     [0.0, 0.0, 1.0]],
    [[1.0, 0.0, 0.0],
     [0.0, 0.9, -0.436],
     [0.0, 0.436, 0.9]]
], dtype=torch.float32)

ARGS = {
    'rotation_matrix': rotation_matrix,
    'eps': 1e-06
}
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

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, 1 are also present in other papers whose code we hold (1 with at least one paper dated earlier, 0 the same month, 0 later) and 7 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
rotation_matrix_to_quaternion 1 other paper
1 dated earlier
2007.11110 (2020-07)
Who Left the Dogs Out? 3D Animal Reconstruction with Expectation Maximization in the Loop

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

Accurate grasping is the key to several robotic tasks including assembly and household robotics. Executing a successful grasp in a cluttered environment requires multiple levels of scene understanding: First, the robot needs to analyze the geometric properties of individual objects to find feasible grasps. These grasps need to be compliant with the local object geometry. Second, for each proposed grasp, the robot needs to reason about the interactions with other objects in the scene. Finally, the robot must compute a collision-free grasp trajectory while taking into account the geometry of the target object. Most grasp detection algorithms directly predict grasp poses in a monolithic fashion, which does not capture the composability of the environment. In this paper, we introduce an end-to-end architecture for object-centric grasping. The method uses pointcloud data from a single arbitrary viewing direction as an input and generates an instance-centric representation for each partially observed object in the scene. This representation is further used for object reconstruction and grasp detection in cluttered table-top scenes. We show the effectiveness of the proposed method by extensively evaluating it against state-of-the-art methods on synthetic datasets, indicating superior performance for grasping and reconstruction. Additionally, we demonstrate real-world applicability by decluttering scenes with varying numbers of objects.

For agents

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

get_harvested_code_for_paper("2401.09939")
get_code_for_paper("2401.09939")
have("2401.09939")

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

curl https://syntology.ai/api/ran/2401.09939.json

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

[![Syntology run record](https://syntology.ai/api/ran/2401.09939.svg)](https://syntology.ai/paper/2401.09939)

Connect an agent — have() is free.