SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2004.06002 · 2020

Dynamic R-CNN: Towards High Quality Object Detection via Dynamic Training

arXiv · PDF · Open in the Atlas

Code that ran

9 of 18 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; 1 honoured the contract we drafted, 1 violated it and 4 have no contract verdict; 3 failures of our instrument; 9 no recorded run), measured 2026-11-21; 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/2004.06002.json · badge: /api/ran/2004.06002.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
hkzhang95/DynamicRCNN canonical (classifier label)
classifier basis: census
3 ran · 1 no recorded run
TangShengqin/DynamicRCNN-mmdet community (classifier label)
classifier basis not recorded on the link
6 ran · 8 no recorded run
FunctionStatusWhere it lives
ensure_rng Ran
ran; our contract draft was wrong, not the code
TangShengqin/DynamicRCNN-mmdet/mmdet/core/bbox/demodata.py
code served (permissive licence) · get_code("8f43d2c26227a141")
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 RandomState. 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

# Test case 1: None (uses global rng)
case_1_rng = None

# Test case 2: integer seed
case_2_rng = 42

# Test case 3: RandomState object
case_3_rng = np.random.RandomState(123)

ARGS = {
    "rng": case_1_rng,
}
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 RandomState — a RandomState output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a RandomState 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
expand_target Ran
ran on a synthesized input
TangShengqin/DynamicRCNN-mmdet/mmdet/core/bbox/bbox_target.py
code served (permissive licence) · get_code("a57ba57f6dde621b")
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 3, 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 batch_size=2, bbox_dim=4
batch_size = 2
num_classes = 3

# bbox_targets: (batch_size, 4) - bounding box regression targets
bbox_targets = torch.randn(batch_size, 4)

# bbox_weights: (batch_size, 4) - weights for each bbox coordinate
bbox_weights = torch.ones(batch_size, 4)

# labels: (batch_size,) - class labels, values should be > 0 for some samples
# Using values in range [0, num_classes) where > 0 means object is present
labels = torch.tensor([0, 1], dtype=torch.long)

ARGS = {
    "bbox_targets": bbox_targets,
    "bbox_weights": bbox_weights,
    "labels": labels,
    "num_classes": num_classes,
}
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 3, 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
ga_loc_target Ran
ran on a synthesized input
TangShengqin/DynamicRCNN-mmdet/mmdet/core/anchor/guided_anchor_target.py
code served (permissive licence) · get_code("2440313353f0c835")
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 torch

torch.manual_seed(0)

# Create ground truth bboxes for 2 images
# Format: [x1, y1, x2, y2] for each bbox
# Image 1: 2 bboxes
gt_bboxes_img1 = torch.tensor([
    [10.0, 10.0, 50.0, 50.0],
    [60.0, 60.0, 100.0, 100.0]
], dtype=torch.float32)

# Image 2: 1 bbox
gt_bboxes_img2 = torch.tensor([
    [20.0, 20.0, 80.0, 80.0]
], dtype=torch.float32)

gt_bboxes_list = [gt_bboxes_img1, gt_bboxes_img2]

# Feature map sizes at different pyramid levels
# Format: (height, width) for each level
featmap_sizes = [
    (64, 64),   # Level 0
    (32, 32),   # Level 1
    (16, 16),   # Level 2
]

# Anchor configuration
anchor_scale = 4
anchor_strides = [4, 8, 16]  # Stride for each pyramid level

# Center and ignore ratios
center_ratio = 0.2
ignore_ratio = 0.5

ARGS = {
    'gt_bboxes_list': gt_bboxes_list,
    'featmap_sizes': featmap_sizes,
    'anchor_scale': anchor_scale,
    'anchor_strides': anchor_strides,
    'center_ratio': center_ratio,
    'ignore_ratio': ignore_ratio,
}
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 — 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
has_valid_annotation Ran
ran, violated the contract we drafted
hkzhang95/DynamicRCNN/dynamic_rcnn/datasets/coco.py
code served (permissive licence) · get_code("2531c930ac17dddf")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Sonnet 4.5, prompt python_fixture_v1) and returned bool, 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 numpy as np

torch.manual_seed(0)

# The function expects 'anno' which is a list of annotation dictionaries
# Each annotation dict can have:
# - 'bbox': a list/array with at least 4 elements [x, y, width, height]
# - 'keypoints': a list where every 3rd element (starting from index 2) is a visibility flag

# Create a valid annotation with keypoints
anno_valid = [
    {
        'bbox': [10, 20, 50, 60],  # valid bbox with width=50, height=60
        'keypoints': [
            100, 200, 2,  # keypoint 1: x, y, visibility=2 (visible)
            110, 210, 1,  # keypoint 2: x, y, visibility=1 (visible)
            120, 220, 2,  # keypoint 3: x, y, visibility=2 (visible)
            130, 230, 0,  # keypoint 4: x, y, visibility=0 (not visible)
            140, 240, 2,  # keypoint 5: x, y, visibility=2 (visible)
            150, 250, 1,  # keypoint 6: x, y, visibility=1 (visible)
            160, 260, 2,  # keypoint 7: x, y, visibility=2 (visible)
            170, 270, 2,  # keypoint 8: x, y, visibility=2 (visible)
            180, 280, 1,  # keypoint 9: x, y, visibility=1 (visible)
            190, 290, 2,  # keypoint 10: x, y, visibility=2 (visible)
            200, 300, 2,  # keypoint 11: x, y, visibility=2 (visible)
        ]
    },
    {
        'bbox': [15, 25, 55, 65],
        'keypoints': [
            105, 205, 1,
            115, 215, 0,
            125, 225, 1,
        ]
    }
]

ARGS = {'anno': anno_valid}
Setup block drafted by
a language model, Claude Sonnet 4.5, prompt python_fixture_v1
Imports
numpy, torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type bool, 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
images_to_levels Ran
ran on a synthesized input
TangShengqin/DynamicRCNN-mmdet/mmdet/core/anchor/anchor_target.py
code served (permissive licence) · get_code("8fa249eba3a0ce57")
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. 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

torch.manual_seed(0)

# The function expects:
# - target: a list of tensors (one per image)
# - num_level_anchors: a list of integers representing anchor counts per level

# Create a list of target tensors (one per image)
# Each tensor has shape (num_anchors,) where num_anchors is the total across all levels
target = [
    torch.randn(7),  # Image 0: 7 total anchors
    torch.randn(7),  # Image 1: 7 total anchors
]

# Define how many anchors per level
# Total anchors = 3 + 4 = 7
num_level_anchors = [3, 4]

ARGS = {
    "target": target,
    "num_level_anchors": num_level_anchors,
}
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 list — a list 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
images_to_levels Ran
ran on a synthesized input
TangShengqin/DynamicRCNN-mmdet/mmdet/core/anchor/point_target.py
code served (permissive licence) · get_code("99bbbfc95abdc86a")
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. 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

torch.manual_seed(0)

# Analyzing the function:
# - target: a list of tensors (one per image) that will be stacked into shape (num_images, total_grid_points)
# - num_level_grids: a list of integers representing grid counts per level
# 
# The function stacks targets, then splits them by level based on num_level_grids

# Create target list - 2 images with features at different grid points
# Total grid points should equal sum of num_level_grids
num_level_grids = [6, 4, 2]  # 3 levels with 6, 4, 2 grid points respectively
total_grid_points = sum(num_level_grids)  # 12

# Create 2 image targets, each with 12 grid point features
target = [
    torch.randn(total_grid_points),  # image 0
    torch.randn(total_grid_points),  # image 1
]

ARGS = {
    "target": target,
    "num_level_grids": num_level_grids,
}
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 list — a list 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
random_boxes Ran
ran, honoured the contract we drafted
TangShengqin/DynamicRCNN-mmdet/mmdet/core/bbox/demodata.py
code served (permissive licence) · get_code("3dc30523833b925a")
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 (5, 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 torch

torch.manual_seed(0)

ARGS = {
    "num": 5,
    "scale": 512,
    "rng": 42,
}
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 2, dtype family float, shape (5, 4) — shape (5, 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
sigmoid_focal_loss_cpu Ran
ran; our contract draft was wrong, not the code
hkzhang95/DynamicRCNN/dynamic_rcnn/det_opr/loss.py
pointer only (licence: MIT); the stored copy get_code serves, from KanghoonYoon/torch-rasgg, is pointer only (licence: NONE) · get_code("426fd02a2beaf826")
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 KanghoonYoon/torch-rasgg, 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.

smooth_l1_loss Ran
ran; our contract draft was wrong, not the code
hkzhang95/DynamicRCNN/dynamic_rcnn/det_opr/loss.py
code served (permissive licence) · get_code("e261fa29066b37e5")
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 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

torch.manual_seed(0)

# Create small tensors for input and target
input_tensor = torch.randn(2, 4)
target_tensor = torch.randn(2, 4)

# Create ARGS dict with all parameters
ARGS = {
    "input": input_tensor,
    "target": target_tensor,
    "beta": 1.0 / 9,
    "size_average": True,
}
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 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
anchor_target Not yet run
unverified (harvested, no recorded run)
TangShengqin/DynamicRCNN-mmdet/mmdet/core/anchor/anchor_target.py
code served (permissive licence) · get_code("6a7ed994a87db611")
How it ran

no invocation recorded.

anchor_target_single Not yet run
unverified (harvested, no recorded run)
TangShengqin/DynamicRCNN-mmdet/mmdet/core/anchor/anchor_target.py
code served (permissive licence) · get_code("1dd00ccb42e850dc")
How it ran

no invocation recorded.

bbox_target Not yet run
unverified (harvested, no recorded run)
TangShengqin/DynamicRCNN-mmdet/mmdet/core/bbox/bbox_target.py
code served (permissive licence) · get_code("34a7030ddea70bdd")
How it ran

no invocation recorded.

bbox_target_single Not yet run
unverified (harvested, no recorded run)
TangShengqin/DynamicRCNN-mmdet/mmdet/core/bbox/bbox_target.py
code served (permissive licence) · get_code("a66c7009b985da0a")
How it ran

no invocation recorded.

calc_region Not yet run
unverified (harvested, no recorded run)
TangShengqin/DynamicRCNN-mmdet/mmdet/core/anchor/guided_anchor_target.py
code served (permissive licence) · get_code("736c7ce8a0faa02a")
How it ran

no invocation recorded.

ga_shape_target Not yet run
unverified (harvested, no recorded run)
TangShengqin/DynamicRCNN-mmdet/mmdet/core/anchor/guided_anchor_target.py
code served (permissive licence) · get_code("e3463fa8772a18a9")
How it ran

no invocation recorded.

load_resnet_c2_format Not yet run
unverified (harvested, no recorded run)
hkzhang95/DynamicRCNN/dynamic_rcnn/basemodels/c2_model_loading.py
code served (permissive licence) · get_code("b99b85e38568b360")
How it ran

no invocation recorded.

point_target Not yet run
unverified (harvested, no recorded run)
TangShengqin/DynamicRCNN-mmdet/mmdet/core/anchor/point_target.py
code served (permissive licence) · get_code("a08ee98e88ad24e7")
How it ran

no invocation recorded.

point_target_single Not yet run
unverified (harvested, no recorded run)
TangShengqin/DynamicRCNN-mmdet/mmdet/core/anchor/point_target.py
code served (permissive licence) · get_code("b660461f38b4382b")
How it ran

no invocation recorded.

Also present in other papers' code

Of 18 samples lifted from this paper's linked repositories, 18 carry a function identity and 0 do not. Of the 18 distinct functions with an identity, 17 are also present in other papers whose code we hold (17 with at least one paper dated earlier, 2 the same month, 13 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
smooth_l1_loss 16 other papers
6 dated earlier, 1 the same month, 9 later
1703.06870 (2017-03)
Mask R-CNN
1903.00241 (2019-03)
Mask Scoring R-CNN
1904.01169 (2019-04)
Res2Net: A New Multi-scale Backbone Architecture
and 13 more not named here
sigmoid_focal_loss_cpu 10 other papers
4 dated earlier, 1 the same month, 5 later
1803.03243 (2018-03)
Domain Adaptive Faster R-CNN for Object Detection in the Wild
1904.01169 (2019-04)
Res2Net: A New Multi-scale Backbone Architecture
1904.01355 (2019-04)
FCOS: Fully Convolutional One-Stage Object Detection
and 7 more not named here
images_to_levels 6 other papers
5 dated earlier, 1 later
1903.00621 (2019-03)
Feature Selective Anchor-Free Module for Single-Shot Object Detection
1906.07155 (2019-06)
MMDetection: Open MMLab Detection Toolbox and Benchmark
1907.03958 (2019-07)
Attentive CT Lesion Detection Using Deep Pyramid Inference with Multi-Scale Booster
and 3 more not named here
anchor_target 4 other papers
3 dated earlier, 1 later
1903.00621 (2019-03)
Feature Selective Anchor-Free Module for Single-Shot Object Detection
1906.07155 (2019-06)
MMDetection: Open MMLab Detection Toolbox and Benchmark
2003.10152 (2020-03)
SOLOv2: Dynamic and Fast Instance Segmentation
and 1 more not named here
calc_region 4 other papers
3 dated earlier, 1 later
1903.00621 (2019-03)
Feature Selective Anchor-Free Module for Single-Shot Object Detection
1906.07155 (2019-06)
MMDetection: Open MMLab Detection Toolbox and Benchmark
2003.10152 (2020-03)
SOLOv2: Dynamic and Fast Instance Segmentation
and 1 more not named here
ga_loc_target 4 other papers
3 dated earlier, 1 later
1903.00621 (2019-03)
Feature Selective Anchor-Free Module for Single-Shot Object Detection
1906.07155 (2019-06)
MMDetection: Open MMLab Detection Toolbox and Benchmark
2003.10152 (2020-03)
SOLOv2: Dynamic and Fast Instance Segmentation
and 1 more not named here
ga_shape_target 4 other papers
3 dated earlier, 1 later
1903.00621 (2019-03)
Feature Selective Anchor-Free Module for Single-Shot Object Detection
1906.07155 (2019-06)
MMDetection: Open MMLab Detection Toolbox and Benchmark
2003.10152 (2020-03)
SOLOv2: Dynamic and Fast Instance Segmentation
and 1 more not named here
point_target 4 other papers
3 dated earlier, 1 later
1903.00621 (2019-03)
Feature Selective Anchor-Free Module for Single-Shot Object Detection
1906.07155 (2019-06)
MMDetection: Open MMLab Detection Toolbox and Benchmark
2003.10152 (2020-03)
SOLOv2: Dynamic and Fast Instance Segmentation
and 1 more not named here
point_target_single 4 other papers
3 dated earlier, 1 later
1903.00621 (2019-03)
Feature Selective Anchor-Free Module for Single-Shot Object Detection
1906.07155 (2019-06)
MMDetection: Open MMLab Detection Toolbox and Benchmark
2003.10152 (2020-03)
SOLOv2: Dynamic and Fast Instance Segmentation
and 1 more not named here
images_to_levels 3 other papers
2 dated earlier, 1 later
1906.07155 (2019-06)
MMDetection: Open MMLab Detection Toolbox and Benchmark
2003.10152 (2020-03)
SOLOv2: Dynamic and Fast Instance Segmentation
2108.07020 (2021-08)
Towards Real-World Prohibited Item Detection: A Large-Scale X-ray Benchmark
anchor_target_single 2 other papers
1 dated earlier, 1 later
1906.07155 (2019-06)
MMDetection: Open MMLab Detection Toolbox and Benchmark
2108.07020 (2021-08)
Towards Real-World Prohibited Item Detection: A Large-Scale X-ray Benchmark
ensure_rng 2 other papers
1 dated earlier, 1 later
1906.07155 (2019-06)
MMDetection: Open MMLab Detection Toolbox and Benchmark
2303.12396 (2023-03)
Rigidity-Aware Detection for 6D Object Pose Estimation
expand_target 2 other papers
2 dated earlier
1906.07155 (2019-06)
MMDetection: Open MMLab Detection Toolbox and Benchmark
2003.13042 (2020-03)
Omni-sourced Webly-supervised Learning for Video Recognition
random_boxes 2 other papers
1 dated earlier, 1 later
1906.07155 (2019-06)
MMDetection: Open MMLab Detection Toolbox and Benchmark
2303.12396 (2023-03)
Rigidity-Aware Detection for 6D Object Pose Estimation
bbox_target 1 other paper
1 dated earlier
1906.07155 (2019-06)
MMDetection: Open MMLab Detection Toolbox and Benchmark
bbox_target_single 1 other paper
1 dated earlier
1906.07155 (2019-06)
MMDetection: Open MMLab Detection Toolbox and Benchmark
has_valid_annotation 1 other paper
1 dated earlier
1904.01355 (2019-04)
FCOS: Fully Convolutional One-Stage Object Detection

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 2026-11-21.

Repositories linked to this paper

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

Abstract

Although two-stage object detectors have continuously advanced the state-of-the-art performance in recent years, the training process itself is far from crystal. In this work, we first point out the inconsistency problem between the fixed network settings and the dynamic training procedure, which greatly affects the performance. For example, the fixed label assignment strategy and regression loss function cannot fit the distribution change of proposals and thus are harmful to training high quality detectors. Consequently, we propose Dynamic R-CNN to adjust the label assignment criteria (IoU threshold) and the shape of regression loss function (parameters of SmoothL1 Loss) automatically based on the statistics of proposals during training. This dynamic design makes better use of the training samples and pushes the detector to fit more high quality samples. Specifically, our method improves upon ResNet-50-FPN baseline with 1.9% AP and 5.5% AP$_{90}$ on the MS COCO dataset with no extra overhead. Codes and models are available at https://github.com/hkzhang95/DynamicRCNN.

For agents

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

get_harvested_code_for_paper("2004.06002")
get_code_for_paper("2004.06002")
have("2004.06002")

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

curl https://syntology.ai/api/ran/2004.06002.json

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

[![Syntology run record](https://syntology.ai/api/ran/2004.06002.svg)](https://syntology.ai/paper/2004.06002)

Connect an agent — have() is free.