SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2307.01197 · 2023

Segment Anything Meets Point Tracking

arXiv · PDF · Open in the Atlas

Code that ran

7 of 12 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 7 have no contract verdict; 0 failures of our instrument; 5 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/2307.01197.json · badge: /api/ran/2307.01197.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
syscv/sam-pt canonical (classifier label)
classifier basis: census
7 ran · 5 no recorded run
FunctionStatusWhere it lives
add_mask_to_frame Ran
ran on a synthesized input
syscv/sam-pt/sam_pt/utils/util.py
code served (permissive licence) · get_code("7e20bee8c9ab288b")
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 3, dtype family int, shape (4, 6, 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 numpy as np
import cv2

# Create a sample frame (3D array with 3 color channels)
# Using shape (4, 6, 3) - height=4, width=6, channels=3
frame = np.random.randint(0, 256, size=(4, 6, 3), dtype=np.uint8)

# Create a sample mask (2D array)
# Using shape (4, 6) - matching frame height and width
mask = np.random.randint(0, 2, size=(4, 6), dtype=np.uint8)

# Define color as RGB tuple
color = (0.5, 0.7, 0.3)

# Define alpha blending value
alpha = 0.2

ARGS = {
    "frame": frame,
    "mask": mask,
    "color": color,
    "alpha": alpha
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
cv2, numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type ndarray, ndim 3, dtype family int, shape (4, 6, 3) — shape (4, 6, 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
batchify_list Ran
ran on a synthesized input
syscv/sam-pt/sam_pt/utils/util.py
code served (permissive licence) · get_code("1776176a97dd455a")
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 math
import torch

# Set seed for reproducibility
torch.manual_seed(0)

# Define a simple post-processing function that converts a list of integers to a tensor
def simple_post_processing(batch):
    """Convert a batch (list) to a tensor"""
    return torch.tensor(batch, dtype=torch.float32)

# Create sample datapoints
datapoints = list(range(10))  # [0, 1, 2, 3, 4, 5, 6, 7, 8, 9]

# Define arguments
ARGS = {
    "datapoints": datapoints,
    "batch_size": 3,
    "batch_post_processing_fn": simple_post_processing
}
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
extract_random_mask_points Ran
ran on a synthesized input
syscv/sam-pt/sam_pt/utils/query_points.py
code served (permissive licence) · get_code("082bcd7cae968600")
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, 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 torch

torch.manual_seed(0)

# Create a simple binary mask tensor with shape (height, width)
# The mask should have some 1s and some 0s
mask = torch.zeros((8, 6), dtype=torch.float32)
mask[1:4, 2:5] = 1.0  # Set a rectangular region to 1

# Number of points to select
n_points_to_select = 5

ARGS = {
    "mask": mask,
    "n_points_to_select": n_points_to_select,
}
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 (5, 2) — shape (5, 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_linear_schedule_with_warmup Ran
ran on a synthesized input
syscv/sam-pt/sam_pt/utils/util.py
code served (permissive licence) · get_code("9d9402c814a980c3")
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 LambdaLR. 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
import torch.optim as optim

torch.manual_seed(0)

# Create a simple model
model = nn.Linear(4, 3)

# Create an optimizer
optimizer = optim.SGD(model.parameters(), lr=0.1)

# Define the arguments
ARGS = {
    "optimizer": optimizer,
    "num_warmup_steps": 100,
    "num_training_steps": 1000,
    "last_epoch": -1,
}
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 LambdaLR — a LambdaLR output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a LambdaLR 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
interp Ran
ran on a synthesized input
syscv/sam-pt/sam_pt/point_tracker/tapnet/tapnet_model.py
code served (permissive licence) · get_code("e5e6908904107ff9")
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 ArrayImpl, ndim 1, dtype family float, shape (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 jax
import jax.numpy as jnp
import chex

# Set seed for reproducibility
jax.config.update('jax_platform_name', 'cpu')

# Create a small grid of features to be interpolated
# Shape: [height, width]
x = jnp.arange(12, dtype=jnp.float32).reshape(4, 3)

# Create interpolation points
# Test case 1: 2D points [y, x] in pixel coordinates
y_2d = jnp.array([[1.5, 1.2], [2.3, 0.8]], dtype=jnp.float32)

# Test case 2: 3D points [z, y, x] in frame coordinates
y_3d = jnp.array([[0.5, 1.5, 1.2], [1.2, 2.3, 0.8]], dtype=jnp.float32)

# Use the 2D case for the main ARGS
ARGS = {
    "x": x,
    "y": y_2d,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
chex, jax — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type ArrayImpl, ndim 1, dtype family float, shape (2,) — shape (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
soft_argmax_heatmap Ran
ran on a synthesized input
syscv/sam-pt/sam_pt/point_tracker/tapnet/tapnet_model.py
code served (permissive licence) · get_code("2d2918c5490472b1")
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 ArrayImpl, ndim 1, dtype family float, shape (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 jax.numpy as jnp
import chex

# Create a valid heatmap: shape [height, width] with positive values summing to 1
height, width = 4, 3
heatmap = jnp.array([
    [0.1, 0.2, 0.15],
    [0.15, 0.25, 0.05],
    [0.08, 0.12, 0.07],
    [0.02, 0.0, 0.0]
], dtype=jnp.float32)

# Normalize to ensure it sums to 1 (already close, but make it exact)
heatmap = heatmap / jnp.sum(heatmap)

ARGS = {
    "softmax_val": heatmap,
    "threshold": 5
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
chex, jax — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type ArrayImpl, ndim 1, dtype family float, shape (2,) — shape (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
visualize_pred_against_gt_mask Ran
ran on a synthesized input
syscv/sam-pt/sam_pt/modeling/sam_pt_interactive.py
code served (permissive licence) · get_code("c207e26ec20cc1e3")
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 3, dtype family int, shape (64, 64, 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
import numpy as np

torch.manual_seed(0)

# Create small tensors for the visualization
# rgb: (3, H, W) - 3 channel image
H, W = 32, 32
rgb = torch.rand(3, H, W)

# pred_mask: (H, W) - predicted binary mask
pred_mask = torch.randint(0, 2, (H, W)).float()

# gt_mask: (H, W) - ground truth binary mask
gt_mask = torch.randint(0, 2, (H, W)).float()

# point_vis: (N,) - visibility of points (1 or 0)
N = 4
point_vis = torch.tensor([1, 1, 0, 1])

# point_coords: (N, 2) - (x, y) coordinates of points
point_coords = torch.tensor([
    [5.0, 10.0],
    [15.0, 20.0],
    [8.0, 12.0],
    [25.0, 28.0]
])

# point_labels: (N,) - label for each point (1 for positive, 0 for negative)
point_labels = torch.tensor([1, 1, 0, 1])

# debug_text: optional string for debug information
debug_text = "Debug Info"

# plot: whether to display the plot
plot = False

ARGS = {
    "rgb": rgb,
    "pred_mask": pred_mask,
    "gt_mask": gt_mask,
    "point_vis": point_vis,
    "point_coords": point_coords,
    "point_labels": point_labels,
    "debug_text": debug_text,
    "plot": plot,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
cv2, matplotlib, numpy, torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type ndarray, ndim 3, dtype family int, shape (64, 64, 3) — shape (64, 64, 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_channels_mixer Not yet run
unverified (harvested, no recorded run)
syscv/sam-pt/sam_pt/point_tracker/tapir/tapir_model.py
code served (permissive licence) · get_code("fa8999c2e2fb539f")
How it ran

no invocation recorded.

depthwise_conv_residual Not yet run
unverified (harvested, no recorded run)
syscv/sam-pt/sam_pt/point_tracker/tapir/tapir_model.py
code served (permissive licence) · get_code("6d1df288856f878c")
How it ran

no invocation recorded.

heatmaps_to_points Not yet run
unverified (harvested, no recorded run)
syscv/sam-pt/sam_pt/point_tracker/tapnet/tapnet_model.py
code served (permissive licence) · get_code("3c5028c1ec6827e1")
How it ran

no invocation recorded.

layernorm Not yet run
unverified (harvested, no recorded run)
syscv/sam-pt/sam_pt/point_tracker/tapir/tapir_model.py
pointer only (licence: Apache-2.0); the stored copy get_code serves, from chopardda/GMA, is pointer only (licence: NONE) · get_code("64c885d6b9f4fd57")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (Apache-2.0), but get_code serves the stored copy, first harvested from chopardda/GMA, 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. No invocation recorded.

visualize_int Not yet run
unverified (harvested, no recorded run)
syscv/sam-pt/sam_pt/modeling/sam_pt_interactive.py
code served (permissive licence) · get_code("bd92506953233c24")
How it ran

no invocation recorded.

Also present in other papers' code

Of 12 samples lifted from this paper's linked repositories, 12 carry a function identity and 0 do not. Of the 12 distinct functions with an identity, 3 are also present in other papers whose code we hold (2 with at least one paper dated earlier, 0 the same month, 3 later) and 9 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
conv_channels_mixer 3 other papers
1 dated earlier, 2 later
2211.03726 (2022-11)
TAP-Vid: A Benchmark for Tracking Any Point in a Video
2411.09821 (2024-11)
Towards Scalable Newborn Screening: Automated General Movement Assessment in Uncontrolled Settings
2501.12392 (2025-01)
Learning segmentation from point trajectories
depthwise_conv_residual 2 other papers
1 dated earlier, 1 later
2211.03726 (2022-11)
TAP-Vid: A Benchmark for Tracking Any Point in a Video
2411.09821 (2024-11)
Towards Scalable Newborn Screening: Automated General Movement Assessment in Uncontrolled Settings
layernorm 1 other paper
1 later
2411.09821 (2024-11)
Towards Scalable Newborn Screening: Automated General Movement Assessment in Uncontrolled Settings

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 Segment Anything Model (SAM) has established itself as a powerful zero-shot image segmentation model, enabled by efficient point-centric annotation and prompt-based models. While click and brush interactions are both well explored in interactive image segmentation, the existing methods on videos focus on mask annotation and propagation. This paper presents SAM-PT, a novel method for point-centric interactive video segmentation, empowered by SAM and long-term point tracking. SAM-PT leverages robust and sparse point selection and propagation techniques for mask generation. Compared to traditional object-centric mask propagation strategies, we uniquely use point propagation to exploit local structure information agnostic to object semantics. We highlight the merits of point-based tracking through direct evaluation on the zero-shot open-world Unidentified Video Objects (UVO) benchmark. Our experiments on popular video object segmentation and multi-object segmentation tracking benchmarks, including DAVIS, YouTube-VOS, and BDD100K, suggest that a point-based segmentation tracker yields better zero-shot performance and efficient interactions. We release our code that integrates different point trackers and video segmentation benchmarks at https://github.com/SysCV/sam-pt.

For agents

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

get_harvested_code_for_paper("2307.01197")
get_code_for_paper("2307.01197")
have("2307.01197")

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

curl https://syntology.ai/api/ran/2307.01197.json

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

[![Syntology run record](https://syntology.ai/api/ran/2307.01197.svg)](https://syntology.ai/paper/2307.01197)

Connect an agent — have() is free.