SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 1807.09190 · 2018

PReMVOS: Proposal-generation, Refinement and Merging for Video Object Segmentation

arXiv · PDF · Open in the Atlas

Code that ran

12 of 15 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 8 have no contract verdict; 4 failures of our instrument; 3 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/1807.09190.json · badge: /api/ran/1807.09190.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
AyaLotfy/PReMVOS-trial community (classifier label)
classifier basis not recorded on the link
7 ran · 2 no recorded run
copy not recorded community
not in the paper's declared repositories
3 ran · 0 no recorded run
VisualComputingInstitute/mots_trackingonly_tools community (classifier label)
classifier basis not recorded on the link
2 ran · 1 no recorded run
FunctionStatusWhere it lives
average_gradients Ran
ran; our contract draft was wrong, not the code
AyaLotfy/PReMVOS-trial/code/ReID_net/Util.py
code served (permissive licence) · get_code("d82c26b8b96a063f")
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, ndim 4, 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 tensorflow as tf
import numpy as np

# Create tower_grads: List of lists of (gradient, variable) tuples
# Structure: outer list is over gradients, inner list is over towers

# Let's create 2 towers with 3 shared variables each
num_towers = 2
num_variables = 3

tower_grads = []

# For each variable, we have a list of (gradient, variable) tuples across towers
for var_idx in range(num_variables):
    grad_and_vars_per_tower = []
    
    # Create a variable (use a simple tensor as a stand-in)
    variable = tf.Variable(tf.ones([4, 3], dtype=tf.float32), trainable=True)
    
    # For each tower, create a gradient and pair it with the variable
    for tower_idx in range(num_towers):
        # Create a gradient tensor with shape [4, 3]
        gradient = tf.constant(
            np.random.randn(4, 3).astype(np.float32),
            dtype=tf.float32
        )
        grad_and_vars_per_tower.append((gradient, variable))
    
    tower_grads.append(grad_and_vars_per_tower)

ARGS = {
    "tower_grads": tower_grads
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, tensorflow — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type list, ndim 4, 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
bbox2 Ran
ran on a synthesized input
AyaLotfy/PReMVOS-trial/code/ReID_net/prepare_input.py
code served (permissive licence) · get_code("3d9a814bfc32150c")
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 1, dtype family int; 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 numpy as np

# The function bbox2 takes an image array and finds the bounding box
# of non-zero elements. It uses np.any() along axes and np.where()
# to find row and column indices.

# Create a small image with some non-zero elements
# Shape: (5, 4) - height=5, width=4
img = np.zeros((5, 4), dtype=np.uint8)
# Add some non-zero values in the middle
img[1:4, 1:3] = 1

ARGS = {
    "img": img
}
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, ndim 1, dtype family int — 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
clip_gradients Ran
ran on a synthesized input
AyaLotfy/PReMVOS-trial/code/ReID_net/Util.py
code served (permissive licence) · get_code("fe9c62a98e81d5ca")
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 tensorflow as tf
import numpy as np

# Create a simple model to generate gradients
class SimpleModel(tf.keras.Model):
    def __init__(self):
        super().__init__()
        self.dense1 = tf.keras.layers.Dense(4)
        self.dense2 = tf.keras.layers.Dense(3)
    
    def call(self, x):
        x = self.dense1(x)
        x = self.dense2(x)
        return x

model = SimpleModel()

# Create dummy input and compute gradients
x = tf.constant(np.random.randn(2, 5).astype(np.float32))
y = tf.constant(np.random.randn(2, 3).astype(np.float32))

with tf.GradientTape() as tape:
    logits = model(x)
    loss = tf.reduce_mean(tf.square(logits - y))

grads = tape.gradient(loss, model.trainable_variables)

# Create list of (gradient, variable) tuples
grads_and_vars = list(zip(grads, model.trainable_variables))

ARGS = {
    "grads": grads_and_vars,
    "threshold": 1.0
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, tensorflow — 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
compute_binary_ious_tf Ran
ran on a synthesized input
AyaLotfy/PReMVOS-trial/code/ReID_net/Measures.py
code served (permissive licence) · get_code("68d695364fd9f0ca")
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 float32, 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 numpy

# Analyze the function:
# - compute_binary_ious_tf takes `targets` and `outputs`
# - It zips them together and calls compute_iou_for_binary_segmentation on each pair
# - compute_iou_for_binary_segmentation expects numpy arrays with binary values (0 or 1)
# - It performs logical operations and sum on them

# Create small binary segmentation masks
# targets: list of numpy arrays with binary values
# outputs: list of numpy arrays with binary values

targets = [
    numpy.array([[1, 0, 1, 0], [0, 1, 0, 1]], dtype=numpy.int32),
    numpy.array([[0, 1, 0, 1], [1, 0, 1, 0]], dtype=numpy.int32)
]

outputs = [
    numpy.array([[1, 0, 0, 0], [0, 1, 1, 1]], dtype=numpy.int32),
    numpy.array([[0, 1, 0, 0], [1, 0, 0, 0]], dtype=numpy.int32)
]

ARGS = {
    "targets": targets,
    "outputs": outputs
}
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 float32, 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
compute_ious Ran
ran on a synthesized input
AyaLotfy/PReMVOS-trial/code/ReID_net/Interface.py
code served (permissive licence) · get_code("b92ab02e58b761db")
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 ndarray, ndim 2, dtype family float, shape (2, 2) 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

# The function computes Intersection over Union (IoU) between bounding boxes
# x: shape (N, 4) - N bounding boxes in format [x0, x1, y0, y1]
# y: shape (M, 4) - M bounding boxes in format [x0, x1, y0, y1]
# Returns: shape (N, M) - IoU matrix

# Create sample bounding boxes
# Format: [x_min, x_max, y_min, y_max]
x = numpy.array([
    [0, 2, 0, 2],      # box 1: area = 4
    [1, 3, 1, 3],      # box 2: area = 4
], dtype=numpy.float32)

y = numpy.array([
    [0.5, 2.5, 0.5, 2.5],  # box 3: area = 4
    [2, 4, 2, 4],          # box 4: area = 4 (no overlap with first)
], dtype=numpy.float32)

ARGS = {
    "x": x,
    "y": y,
}
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 ndarray, ndim 2, dtype family float, shape (2, 2) — shape (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
conv Ran
ran; our contract draft was wrong, not the code
this paper's copy was not recorded; identical code first harvested from yanqi1811/PWC-Net
pointer only; the stored copy get_code serves, from yanqi1811/PWC-Net, is pointer only (licence: NOASSERTION) · get_code("2d0cae2c772e2af4")
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.

deconv Ran
ran; our contract draft was wrong, not the code
this paper's copy was not recorded; identical code first harvested from yanqi1811/PWC-Net
pointer only; the stored copy get_code serves, from yanqi1811/PWC-Net, is pointer only (licence: NOASSERTION) · get_code("dc4fb7628379691e")
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.

detection_extractor Ran
ran on a synthesized input
AyaLotfy/PReMVOS-trial/code/ReID_net/Interface.py
code served (permissive licence) · get_code("8a0a002fa0a8e01b")
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 function. 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

# Stand-in for TensorFlow session
class MockSession:
    def run(self, fetches, feed_dict=None):
        # Return mock outputs matching the expected structure
        # det_boxes: [batch, num_detections, 4]
        # det_scores: [batch, num_detections]
        # det_classes: [batch, num_detections]
        # num_detections: [batch]
        batch_size = 2
        num_det = 5
        det_boxes = [[0.1, 0.2, 0.3, 0.4] for _ in range(num_det)]
        det_scores = [0.9, 0.85, 0.8, 0.75, 0.7]
        det_classes = [1, 2, 1, 3, 2]
        num_detections = 5
        return [det_boxes, det_scores, det_classes, num_detections]

# Stand-in for output_layer with outputs attribute
class MockOutputLayer:
    def __init__(self):
        self.outputs = [self._create_output_tuple()]
    
    def _create_output_tuple(self):
        # Tuple of (det_boxes, det_scores, reid, det_classes, num_detections)
        # These are TensorFlow tensor references, not actual values
        det_boxes = "det_boxes_tensor"
        det_scores = "det_scores_tensor"
        reid = "reid_tensor"
        det_classes = "det_classes_tensor"
        num_detections = "num_detections_tensor"
        return (det_boxes, det_scores, reid, det_classes, num_detections)

# Stand-in for valid_data
class MockValidData:
    def __init__(self):
        self.img_filename_placeholder = "img_filename_placeholder"

# Stand-in for engine
class MockEngine:
    def __init__(self):
        self.session = MockSession()
        self.valid_data = MockValidData()

# Create mock instances
engine = MockEngine()
output_layer = MockOutputLayer()

ARGS = {
    "engine": engine,
    "output_layer": output_layer
}
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 function — a function output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a function 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
extract Ran
ran on a synthesized input
AyaLotfy/PReMVOS-trial/code/MergeTrack/refinement_net_functions.py
code served (permissive licence) · get_code("1ea937ed7f4c6dbb")
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 1, dtype family float, shape (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
import numpy as np

torch.manual_seed(0)

# The function extracts a value from a dict called 'extractions'
# where each value is a list containing one tensor
# that tensor has shape (1, ...) and we extract element [0]

# Create a small tensor with shape (1, 4) - batch size 1, feature dim 4
small_tensor = torch.randn(1, 4)

# Create the extractions dict with a key mapping to a list containing one tensor
extractions = {
    "feature_key": [small_tensor],
    "another_key": [torch.randn(1, 3)]
}

key = "feature_key"

ARGS = {
    "key": key,
    "extractions": extractions
}
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 1, dtype family float, shape (4,) — shape (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
load_detections_for_seq Ran
ran on a synthesized input
VisualComputingInstitute/mots_trackingonly_tools/io_utils.py
code served (permissive licence) · get_code("093913a1b1c9347a")
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 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 tempfile
import os

# Create a temporary detection file
temp_dir = tempfile.mkdtemp()
det_file = os.path.join(temp_dir, "detections.txt")

# Write sample detection data
# Format: frame_id x1 y1 x2 y2 confidence class_id height width rle_str
with open(det_file, 'w') as f:
    # Frame 0 with detection
    f.write("0 10.5 20.3 100.2 150.7 0.95 1 480 640 eJzj4pIIAgAE5AFy\n")
    # Frame 1 with detection
    f.write("1 15.0 25.0 105.0 155.0 0.87 2 480 640 eJzj4pIIAgAE5AFy\n")
    # Frame 0 with another detection
    f.write("0 200.0 250.0 300.0 350.0 0.92 1 480 640 eJzj4pIIAgAE5AFy\n")

ARGS = {
    "det_fn": det_file,
    "decode_masks": False,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
pycocotools — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type tuple, ndim 1, 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
load_seqmap Ran
ran on a synthesized input
VisualComputingInstitute/mots_trackingonly_tools/io_utils.py
code served (permissive licence) · get_code("24124b92fe9f5ae2")
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 OrderedDict. 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 tempfile
import os
from collections import OrderedDict

class Sequence:
    def __init__(self, id_, end_time, start_time=0):
        self.id_ = id_
        self.end_time = end_time
        self.start_time = start_time

def load_seqmap(seqmap_filename):
    seqs = OrderedDict()
    with open(seqmap_filename) as f:
        for l in f:
            sp = l.split()
            assert len(sp) == 4
            id_ = sp[0]
            start_time = int(sp[2])
            end_time = int(sp[3])
            seq = Sequence(id_, start_time, end_time)
            seqs[id_] = seq
    return seqs

# Create a temporary seqmap file
temp_dir = tempfile.mkdtemp()
seqmap_file = os.path.join(temp_dir, "seqmap.txt")

with open(seqmap_file, 'w') as f:
    f.write("seq_001 irrelevant_name 0 100\n")
    f.write("seq_002 irrelevant_name 50 150\n")
    f.write("seq_003 irrelevant_name 10 200\n")

ARGS = {
    "seqmap_filename": seqmap_file
}
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 OrderedDict — a OrderedDict output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a OrderedDict 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
predict_flow Ran
ran; our contract draft was wrong, not the code
this paper's copy was not recorded; identical code first harvested from yanqi1811/PWC-Net
pointer only; the stored copy get_code serves, from yanqi1811/PWC-Net, is pointer only (licence: NOASSERTION) · get_code("c3f23a008d953af5")
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.

create_confusion_matrix Not yet run
unverified (harvested, no recorded run)
AyaLotfy/PReMVOS-trial/code/ReID_net/Measures.py
code served (permissive licence) · get_code("c463048a0a86251e")
How it ran

no invocation recorded.

get_average_precision Not yet run
unverified (harvested, no recorded run)
AyaLotfy/PReMVOS-trial/code/ReID_net/Measures.py
code served (permissive licence) · get_code("9746592ab02ab201")
How it ran

no invocation recorded.

load_detections Not yet run
unverified (harvested, no recorded run)
VisualComputingInstitute/mots_trackingonly_tools/io_utils.py
code served (permissive licence) · get_code("99b934888578c83a")
How it ran

no invocation recorded.

Also present in other papers' code

Of 15 samples lifted from this paper's linked repositories, 15 carry a function identity and 0 do not. Of the 15 distinct functions with an identity, 4 are also present in other papers whose code we hold (4 with at least one paper dated earlier, 0 the same month, 4 later) and 11 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
predict_flow 11 other papers
1 dated earlier, 10 later
1709.02371 (2017-09)
PWC-Net: CNNs for Optical Flow Using Pyramid, Warping, and Cost Volume
1810.10510 (2018-10)
Neighbourhood Consensus Networks
1904.00830 (2019-04)
Depth-Aware Video Frame Interpolation
and 8 more not named here
deconv 9 other papers
1 dated earlier, 8 later
1709.02371 (2017-09)
PWC-Net: CNNs for Optical Flow Using Pyramid, Warping, and Cost Volume
1810.10510 (2018-10)
Neighbourhood Consensus Networks
1904.00830 (2019-04)
Depth-Aware Video Frame Interpolation
and 6 more not named here
average_gradients 3 other papers
2 dated earlier, 1 later
1803.03764 (2018-03)
Variance Networks: When Expectation Does Not Meet Your Expectations
1805.01954 (2018-05)
Behavioral Cloning from Observation
2003.03030 (2020-03)
Clean-Label Backdoor Attacks on Video Recognition Models
conv 3 other papers
1 dated earlier, 2 later
1709.02371 (2017-09)
PWC-Net: CNNs for Optical Flow Using Pyramid, Warping, and Cost Volume
1904.00830 (2019-04)
Depth-Aware Video Frame Interpolation
2011.00359 (2020-11)
TartanVO: A Generalizable Learning-based VO

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 address semi-supervised video object segmentation, the task of automatically generating accurate and consistent pixel masks for objects in a video sequence, given the first-frame ground truth annotations. Towards this goal, we present the PReMVOS algorithm (Proposal-generation, Refinement and Merging for Video Object Segmentation). Our method separates this problem into two steps, first generating a set of accurate object segmentation mask proposals for each video frame and then selecting and merging these proposals into accurate and temporally consistent pixel-wise object tracks over a video sequence in a way which is designed to specifically tackle the difficult challenges involved with segmenting multiple objects across a video sequence. Our approach surpasses all previous state-of-the-art results on the DAVIS 2017 video object segmentation benchmark with a J & F mean score of 71.6 on the test-dev dataset, and achieves first place in both the DAVIS 2018 Video Object Segmentation Challenge and the YouTube-VOS 1st Large-scale Video Object Segmentation Challenge.

For agents

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

get_harvested_code_for_paper("1807.09190")
get_code_for_paper("1807.09190")
have("1807.09190")

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

curl https://syntology.ai/api/ran/1807.09190.json

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

[![Syntology run record](https://syntology.ai/api/ran/1807.09190.svg)](https://syntology.ai/paper/1807.09190)

Connect an agent — have() is free.