SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 1711.09577 · 2017

Can Spatiotemporal 3D CNNs Retrace the History of 2D CNNs and ImageNet?

arXiv · PDF · Open in the Atlas

Code that ran

7 of 8 samples (functions and classes) lifted from this paper's linked repositories executed on a synthesized input in Syntology's jail (0 of them constructed an object rather than computing a result; 0 honoured the contract we drafted, 0 violated it and 0 have no contract verdict; 7 failures of our instrument; 1 no recorded run), measured 2026-11-22; repository links from the Papers with Code archive attributed under CC BY-SA, no affiliation; bodies inline only under a permissive licence as GitHub detects it. Not a reproduction of the paper's results.

"Ran" means the function executed on a synthesized input and returned a value. It is not a reproduction of the paper's results. This record, dated, as JSON: /api/ran/1711.09577.json · badge: /api/ran/1711.09577.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
tianhai123/3D-ResNets canonical (classifier label)
classifier basis: classifier_2025
2 ran · 0 no recorded run
copy not recorded community
not in the paper's declared repositories
2 ran · 1 no recorded run
okankop/Efficient-3DCNNs reimplementation (classifier label)
classifier basis: classifier_2025
2 ran · 0 no recorded run
arundhatikurup/3DResnet reimplementation (classifier label)
classifier basis: classifier_2025
1 ran · 0 no recorded run
FunctionStatusWhere it lives
conv1x1x1 Ran
ran; our contract draft was wrong, not the code
this paper's copy was not recorded; identical code first harvested from ziqi-zhang/3D-ResNets-PyTorch
pointer only · get_code("213ccc8fb87ee8d9")
How it ran

The body of this sample is withheld: this paper's copy of the code was not recorded (its link names no repository or source path), so no licence was read for it. 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.

conv3x3x3 Ran
ran; our contract draft was wrong, not the code
this paper's copy was not recorded; identical code first harvested from ziqi-zhang/3D-ResNets-PyTorch
pointer only · get_code("c3b215555357b9d1")
How it ran

The body of this sample is withheld: this paper's copy of the code was not recorded (its link names no repository or source path), so no licence was read for it. 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.

downsample_basic_block Ran
ran; our contract draft was wrong, not the code
tianhai123/3D-ResNets/models/resnet.py
code served (permissive licence) · get_code("43c0211ff8dbca5f")
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 5, dtype family float, shape (2, 8, 2, 2, 2) as a later run recorded it. The image it ran in was not recorded. The code mentions cuda, and the harness aliases cuda to the CPU for every sample; whether this run needed that is 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
from torch.autograd import Variable

torch.manual_seed(0)

# Create a small 5D tensor (batch_size=2, channels=4, depth=3, height=3, width=3)
# This matches the expected input for a 3D convolution operation
x = torch.randn(2, 4, 3, 3, 3)

# planes should be greater than the number of input channels (4)
# so that zero padding can be added
planes = 8

# stride for the pooling operation
stride = 2

ARGS = {
    "x": x,
    "planes": planes,
    "stride": stride
}
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 5, dtype family float, shape (2, 8, 2, 2, 2) — shape (2, 8, 2, 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
get_fine_tuning_parameters Ran
ran; our contract draft was wrong, not the code
tianhai123/3D-ResNets/models/resnet.py
code served (permissive licence) · get_code("93801ecf3026ab06")
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 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 torch.nn as nn

torch.manual_seed(0)

# Build a minimal model that has layer0, layer1, layer2, layer3, layer4, and fc
# The function iterates through named_parameters() and checks if module names
# like 'layer0', 'layer1', etc. are in the parameter names
class SimpleModel(nn.Module):
    def __init__(self):
        super().__init__()
        self.layer0 = nn.Linear(4, 3)
        self.layer1 = nn.Linear(3, 3)
        self.layer2 = nn.Linear(3, 3)
        self.layer3 = nn.Linear(3, 3)
        self.layer4 = nn.Linear(3, 2)
        self.fc = nn.Linear(2, 1)
    
    def forward(self, x):
        x = self.layer0(x)
        x = self.layer1(x)
        x = self.layer2(x)
        x = self.layer3(x)
        x = self.layer4(x)
        x = self.fc(x)
        return x

model = SimpleModel()

# Test with ft_begin_index = 0 (should return all parameters)
ft_begin_index_0 = 0

# Test with ft_begin_index = 2 (should fine-tune layer2, layer3, layer4, fc)
ft_begin_index_2 = 2

ARGS = {
    "model": model,
    "ft_begin_index": ft_begin_index_2
}
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 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_fine_tuning_parameters Ran
ran; our contract draft was wrong, not the code
arundhatikurup/3DResnet/models/resnet.py
code served (permissive licence) · get_code("8d57521c7b52b777")
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 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 torch.nn as nn

torch.manual_seed(0)

# Build a minimal model that has named_parameters with layer0-4 and fc
class SimpleModel(nn.Module):
    def __init__(self):
        super().__init__()
        self.layer0 = nn.Linear(4, 3)
        self.layer1 = nn.Linear(3, 3)
        self.layer2 = nn.Linear(3, 3)
        self.layer3 = nn.Linear(3, 3)
        self.layer4 = nn.Linear(3, 2)
        self.fc = nn.Linear(2, 1)
    
    def forward(self, x):
        x = self.layer0(x)
        x = self.layer1(x)
        x = self.layer2(x)
        x = self.layer3(x)
        x = self.layer4(x)
        x = self.fc(x)
        return x

model = SimpleModel()

ARGS = {
    'model': model,
    'ft_begin_index': 2
}
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 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_fine_tuning_parameters Ran
ran; our contract draft was wrong, not the code
okankop/Efficient-3DCNNs/models/resnet.py
code served (permissive licence) · get_code("8484994a54d4c487")
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 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 torch.nn as nn

torch.manual_seed(0)

class SimpleModel(nn.Module):
    def __init__(self):
        super().__init__()
        self.fc = nn.Linear(4, 3)
        self.conv = nn.Conv2d(1, 2, 3)
    
    def forward(self, x):
        return self.fc(x)

model = SimpleModel()

ARGS = {
    'model': model,
    'ft_portion': 'last_layer'
}
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 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
load_labels Ran
ran; our contract draft was wrong, not the code
okankop/Efficient-3DCNNs/utils/kinetics_json.py
code served (permissive licence) · get_code("90b941a8c18bcd57")
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 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 pandas as pd
import tempfile
import os

# Create a temporary CSV file with sample data
temp_dir = tempfile.mkdtemp()
csv_path = os.path.join(temp_dir, "train.csv")

# Write sample data to CSV
data = pd.DataFrame({
    'label': ['cat', 'dog', 'cat', 'bird', 'dog', 'cat']
})
data.to_csv(csv_path, index=False)

ARGS = {
    'train_csv_path': csv_path
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1
Imports
pandas — 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
generate_model Not yet run
unverified (harvested, no recorded run)
this paper's copy was not recorded; identical code first harvested from ziqi-zhang/3D-ResNets-PyTorch
pointer only · get_code("392a1ed5b8624559")
How it ran

The body of this sample is withheld: this paper's copy of the code was not recorded (its link names no repository or source path), so no licence was read for it. 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. No invocation recorded.

Also present in other papers' code

Of 8 samples lifted from this paper's linked repositories, 8 carry a function identity and 0 do not. Of the 8 distinct functions with an identity, 7 are also present in other papers whose code we hold (1 with at least one paper dated earlier, 1 the same month, 7 later) and 1 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
conv3x3x3 19 other papers
1 the same month, 18 later
1711.08580 (2017-11)
3D Anisotropic Hybrid Network: Transferring Convolutional Features from 2D Images to 3D Anisotropic Volumes
1908.01373 (2019-08)
Unsupervised Microvascular Image Segmentation Using an Active Contours Mimicking Neural Network
1911.00232 (2019-11)
Multi-Moments in Time: Learning and Interpreting Models for Multi-Action Video Understanding
and 16 more not named here
downsample_basic_block 15 other papers
1 dated earlier, 14 later
1705.07750 (2017-05)
Quo Vadis, Action Recognition? A New Model and the Kinetics Dataset
1908.01373 (2019-08)
Unsupervised Microvascular Image Segmentation Using an Active Contours Mimicking Neural Network
1909.04656 (2019-09)
Video Representation Learning by Dense Predictive Coding
and 12 more not named here
conv1x1x1 6 other papers
6 later · 1 only as an assembly brick
2003.10432 (2020-03)
Atlas: End-to-End 3D Scene Reconstruction from Posed Images
2004.04968 (2020-04)
Would Mega-scale Datasets Further Enhance Spatiotemporal 3D CNNs?
2107.07695 (2021-07)
Self-supervised Representation Learning Framework for Remote Physiological Measurement Using Spatiotemporal Augmentation Loss
and 3 more not named here
get_fine_tuning_parameters 5 other papers
5 later
1908.01373 (2019-08)
Unsupervised Microvascular Image Segmentation Using an Active Contours Mimicking Neural Network
1912.05534 (2019-12)
Why Can't I Dance in the Mall? Learning to Mitigate Scene Bias in Action Recognition
1912.09930 (2019-12)
Something-Else: Compositional Action Recognition with Spatial-Temporal Interaction Networks
and 2 more not named here
generate_model 3 other papers
3 later
2004.04968 (2020-04)
Would Mega-scale Datasets Further Enhance Spatiotemporal 3D CNNs?
2107.07695 (2021-07)
Self-supervised Representation Learning Framework for Remote Physiological Measurement Using Spatiotemporal Augmentation Loss
2108.02183 (2021-08)
Enhancing Self-supervised Video Representation Learning via Multi-level Feature Optimization
get_fine_tuning_parameters 3 other papers
3 later
2008.11516 (2020-08)
Making a Case for 3D Convolutions for Object Segmentation in Videos
2307.16184 (2023-07)
UnIVAL: Unified Model for Image, Video, Audio and Language Tasks
2311.12052 (2023-11)
MagicPose: Realistic Human Poses and Facial Expressions Retargeting with Identity-aware Diffusion
get_fine_tuning_parameters 1 other paper
1 later
2211.04031 (2022-11)
Hilbert Distillation for Cross-Dimensionality Networks

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

Repositories linked to this paper

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

Abstract

The purpose of this study is to determine whether current video datasets have sufficient data for training very deep convolutional neural networks (CNNs) with spatio-temporal three-dimensional (3D) kernels. Recently, the performance levels of 3D CNNs in the field of action recognition have improved significantly. However, to date, conventional research has only explored relatively shallow 3D architectures. We examine the architectures of various 3D CNNs from relatively shallow to very deep ones on current video datasets. Based on the results of those experiments, the following conclusions could be obtained: (i) ResNet-18 training resulted in significant overfitting for UCF-101, HMDB-51, and ActivityNet but not for Kinetics. (ii) The Kinetics dataset has sufficient data for training of deep 3D CNNs, and enables training of up to 152 ResNets layers, interestingly similar to 2D ResNets on ImageNet. ResNeXt-101 achieved 78.4% average accuracy on the Kinetics test set. (iii) Kinetics pretrained simple 3D architectures outperforms complex 2D architectures, and the pretrained ResNeXt-101 achieved 94.5% and 70.2% on UCF-101 and HMDB-51, respectively. The use of 2D CNNs trained on ImageNet has produced significant progress in various tasks in image. We believe that using deep 3D CNNs together with Kinetics will retrace the successful history of 2D CNNs and ImageNet, and stimulate advances in computer vision for videos. The codes and pretrained models used in this study are publicly available. https://github.com/kenshohara/3D-ResNets-PyTorch

For agents

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

get_harvested_code_for_paper("1711.09577")
get_code_for_paper("1711.09577")
have("1711.09577")

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

curl https://syntology.ai/api/ran/1711.09577.json

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

[![Syntology run record](https://syntology.ai/api/ran/1711.09577.svg)](https://syntology.ai/paper/1711.09577)

Connect an agent — have() is free.