SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2203.09694 · CVPR · 2022

Group Contextualization for Video Recognition

Hao Zhang, Chong-Wah Ngo, Xiangnan He, Yanbin Hao

arXiv · PDF · Open in the Atlas

Code that ran

5 of 5 samples (functions and classes) lifted from this paper's linked repositories executed on a synthesized input in Syntology's jail (5 of them constructed an object rather than computing a result; 0 honoured the contract we drafted, 0 violated it and 5 have no contract verdict; 0 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/2203.09694.json · badge: /api/ran/2203.09694.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
haoyanbin918/Group-Contextualization community
Papers with Code archive
5 ran · 0 no recorded run
FunctionStatusWhere it lives
Bottleneck Ran
ran on a synthesized input · constructed an object; no result was computed
haoyanbin918/Group-Contextualization/nets/GC_GST.py
code served (permissive licence) · get_code("681847d2c90069fd")
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 an instance of its own class, Bottleneck: the run constructed the object and did not call it, so no result was computed. 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)

# Create a minimal downsample module that matches what Bottleneck expects
class MinimalDownsample(nn.Module):
    def __init__(self, in_channels, out_channels):
        super().__init__()
        self.conv = nn.Conv3d(in_channels, out_channels, kernel_size=1, bias=False)
        self.bn = nn.BatchNorm3d(out_channels)
    
    def forward(self, x):
        return self.bn(self.conv(x))

# Set up parameters
inplanes = 64
planes = 32
alpha = 2
beta = 2
stride = 1
downsample = MinimalDownsample(inplanes, planes * 4)  # Bottleneck.expansion = 4
use_ef = False
cdiv = 8
loop_id = 0

ARGS = {
    'inplanes': inplanes,
    'planes': planes,
    'alpha': alpha,
    'beta': beta,
    'stride': stride,
    'downsample': downsample,
    'use_ef': use_ef,
    'cdiv': cdiv,
    'loop_id': loop_id
}
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 Bottleneck — the run constructed an instance of its own class and did not call it; no shape is stated — the instrument coerces any output to an array; for a Bottleneck 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
GC_CLLDnb Ran
ran on a synthesized input · constructed an object; no result was computed
haoyanbin918/Group-Contextualization/nets/GC_GST.py
code served (permissive licence) · get_code("bfd0fc32160812fb")
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 an instance of its own class, GC_CLLDnb: the run constructed the object and did not call it, so no result was computed. 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 GC_CLLDnb(nn.Module):

    def __init__(self, inplanes, planes):
        super(GC_CLLDnb, self).__init__()
        self.avg_pool = nn.AdaptiveAvgPool3d(1)
        self.conv = nn.Linear(inplanes, planes, bias=False)
        self.sigmoid = nn.Sigmoid()
        nn.init.normal_(self.conv.weight, 0, 0.001)

    def forward(self, x):
        bn, c, t, h, w = x.size()
        y = self.avg_pool(x).view(bn, c)
        y = self.conv(y)
        y = self.sigmoid(y).view(bn, c, 1, 1, 1)
        x = x * y.expand_as(x)
        return x

inplanes = 4
planes = 3

ARGS = {
    "inplanes": inplanes,
    "planes": planes,
}
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 GC_CLLDnb — the run constructed an instance of its own class and did not call it; no shape is stated — the instrument coerces any output to an array; for a GC_CLLDnb 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
GC_L33Dnb Ran
ran on a synthesized input · constructed an object; no result was computed
haoyanbin918/Group-Contextualization/nets/GC_GST.py
code served (permissive licence) · get_code("ad22890d898cf1b4")
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 an instance of its own class, GC_L33Dnb: the run constructed the object and did not call it, so no result was computed. 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)

# GC_L33Dnb takes inplanes and planes as parameters
# It's an nn.Module that performs a 3D convolution followed by sigmoid gating
# The forward method expects input x with shape (batch, inplanes, depth, height, width)

# Create small distinct dimensions for each parameter
inplanes = 4
planes = 3

ARGS = {
    "inplanes": inplanes,
    "planes": planes,
}
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 GC_L33Dnb — the run constructed an instance of its own class and did not call it; no shape is stated — the instrument coerces any output to an array; for a GC_L33Dnb 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
GC_S23DDnb Ran
ran on a synthesized input · constructed an object; no result was computed
haoyanbin918/Group-Contextualization/nets/GC_GST.py
code served (permissive licence) · get_code("b46bb46ae87e6dd5")
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 an instance of its own class, GC_S23DDnb: the run constructed the object and did not call it, so no result was computed. 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 valid arguments for GC_S23DDnb
ARGS = {
    'inplanes': 4,
    'planes': 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 GC_S23DDnb — the run constructed an instance of its own class and did not call it; no shape is stated — the instrument coerces any output to an array; for a GC_S23DDnb 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
GC_T13Dnb Ran
ran on a synthesized input · constructed an object; no result was computed
haoyanbin918/Group-Contextualization/nets/GC_GST.py
code served (permissive licence) · get_code("b86f2ed4b012b0fa")
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 an instance of its own class, GC_T13Dnb: the run constructed the object and did not call it, so no result was computed. 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)

# GC_T13Dnb expects:
# - inplanes: int (input channels)
# - planes: int (output channels)
# And in forward():
# - x: tensor of shape (bn, c, t, h, w)

ARGS = {
    "inplanes": 4,
    "planes": 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 GC_T13Dnb — the run constructed an instance of its own class and did not call it; no shape is stated — the instrument coerces any output to an array; for a GC_T13Dnb 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

Also present in other papers' code

Of 5 samples lifted from this paper's linked repositories, 5 carry a function identity and 0 do not. Of the 5 distinct functions with an identity, 0 are also present in other papers whose code we hold (0 with at least one paper dated earlier, 0 the same month, 0 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.

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

Learning discriminative representation from the complex spatio-temporal dynamic space is essential for video recognition. On top of those stylized spatio-temporal computational units, further refining the learnt feature with axial contexts is demonstrated to be promising in achieving this goal. However, previous works generally focus on utilizing a single kind of contexts to calibrate entire feature channels and could hardly apply to deal with diverse video activities. The problem can be tackled by using pair-wise spatio-temporal attentions to recompute feature response with cross-axis contexts at the expense of heavy computations. In this paper, we propose an efficient feature refinement method that decomposes the feature channels into several groups and separately refines them with different axial contexts in parallel. We refer this lightweight feature calibration as group contextualization (GC). Specifically, we design a family of efficient element-wise calibrators, i.e., ECal-G/S/T/L, where their axial contexts are information dynamics aggregated from other axes either globally or locally, to contextualize feature channel groups. The GC module can be densely plugged into each residual layer of the off-the-shelf video networks. With little computational overhead, consistent improvement is observed when plugging in GC on different networks. By utilizing calibrators to embed feature with four different kinds of contexts in parallel, the learnt representation is expected to be more resilient to diverse types of activities. On videos with rich temporal variations, empirically GC can boost the performance of 2D-CNN (e.g., TSN and TSM) to a level comparable to the state-of-the-art video networks. Code is available at https://github.com/haoyanbin918/Group-Contextualization.

For agents

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

get_harvested_code_for_paper("2203.09694")
get_code_for_paper("2203.09694")
have("2203.09694")

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

curl https://syntology.ai/api/ran/2203.09694.json

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

[![Syntology run record](https://syntology.ai/api/ran/2203.09694.svg)](https://syntology.ai/paper/2203.09694)

Connect an agent — have() is free.