SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2407.15420 · ECCV · 2024

Local All-Pair Correspondence for Point Tracking

Seungryong Kim, Jisu Nam, Honggyu An, Seokju Cho, Jiahui Huang, Joon-Young Lee

arXiv · PDF · Open in the Atlas

Code that ran

8 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; 2 honoured the contract we drafted, 0 violated it and 0 have no contract verdict; 6 failures of our instrument; 4 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/2407.15420.json · badge: /api/ran/2407.15420.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
cvlab-kaist/locotrack canonical (classifier label)
classifier basis: census
8 ran · 4 no recorded run
FunctionStatusWhere it lives
add_weight_decay Ran
ran; our contract draft was wrong, not the code
cvlab-kaist/locotrack/locotrack/optimizers.py
pointer only (licence: Apache-2.0); the stored copy get_code serves, from chopardda/GMA, is pointer only (licence: NONE) · get_code("1b97ac62c8bb2152")
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. A run is recorded.

compute_tapvid_metrics Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
cvlab-kaist/locotrack/locotrack/evaluation_datasets.py
code served (permissive licence) · get_code("8d2089efa5f5af9b")
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 dict. 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

# Set random seed for reproducibility
np.random.seed(0)

# Define small but realistic dimensions
batch_size = 2
num_queries = 4
num_frames = 5

# query_points: [batch, num_queries, 3] where format is [t, y, x]
# t should be in range [0, num_frames-1]
query_points = np.array([
    [[0, 10.5, 20.3], [1, 15.2, 25.1], [2, 12.0, 22.5], [0, 18.3, 30.2]],
    [[1, 11.5, 21.3], [2, 14.2, 24.1], [0, 13.0, 23.5], [1, 19.3, 31.2]]
], dtype=np.float32)

# gt_occluded: [batch, num_queries, num_frames] - boolean array
gt_occluded = np.array([
    [[True, False, False, False, False], 
     [False, False, True, False, False],
     [False, False, False, False, True],
     [True, True, False, False, False]],
    [[False, False, False, True, False],
     [True, False, False, False, False],
     [False, False, False, False, False],
     [False, True, True, False, False]]
], dtype=bool)

# gt_tracks: [batch, num_queries, num_frames, 2] format is [x, y]
gt_tracks = np.random.randn(batch_size, num_queries, num_frames, 2).astype(np.float32) * 10 + 50

# pred_occluded: [batch, num_queries, num_frames] - boolean array
pred_occluded = np.array([
    [[True, False, False, False, False], 
     [False, False, True, False, False],
     [False, False, False, False, False],
     [True, True, False, False, True]],
    [[False, False, False, True, False],
     [True, False, False, False, False],
     [False, False, False, False, False],
     [False, True, True, False, False]]
], dtype=bool)

# pred_tracks: [batch, num_queries, num_frames, 2] format is [x, y]
# Add some noise to gt_tracks to simulate predictions
pred_tracks = gt_tracks + np.random.randn(batch_size, num_queries, num_frames, 2).astype(np.float32) * 2

# query_mode: either 'first' or 'strided'
query_mode = 'first'

# get_trackwise_metrics: boolean
get_trackwise_metrics = False

ARGS = {
    'query_points': query_points,
    'gt_occluded': gt_occluded,
    'gt_tracks': gt_tracks,
    'pred_occluded': pred_occluded,
    'pred_tracks': pred_tracks,
    'query_mode': query_mode,
    'get_trackwise_metrics': get_trackwise_metrics
}
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 dict — a dict output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a dict 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
get_alibi_slope Ran
ran, honoured the contract we drafted
cvlab-kaist/locotrack/locotrack/locotrack_model.py
code served (permissive licence) · get_code("e720140bb5365d49")
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 3, dtype family float, shape (8, 1, 1) 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

ARGS = {
    "num_heads": 8
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
jax — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type ArrayImpl, ndim 3, dtype family float, shape (8, 1, 1) — shape (8, 1, 1) 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_relative_positions Ran
ran, honoured the contract we drafted
cvlab-kaist/locotrack/locotrack/locotrack_model.py
code served (permissive licence) · get_code("aa68b05aa2f84e51")
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 2, dtype family int, shape (5, 5) 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

ARGS = {
    "seq_len": 5,
    "reverse": False,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
jax — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type ArrayImpl, ndim 2, dtype family int, shape (5, 5) — shape (5, 5) 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
interp Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
cvlab-kaist/locotrack/locotrack/utils/model_utils.py
code served (permissive licence) · get_code("39fcbdf93dcb4313")
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_default_prng_impl", "threefry2x32")

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

# Create interpolation points [num_points, 2] for 2D case
# Each point is [y, x] in pixel coordinates
y_2d = jnp.array([[1.5, 1.2], [2.3, 0.8]], dtype=jnp.float32)

# Create interpolation points [num_points, 3] for 3D case
# Each point is [z, y, x] where z is frame coordinate
y_3d = jnp.array([[0.5, 1.5, 1.2], [0.2, 2.3, 0.8]], dtype=jnp.float32)

# Test with 2D points
ARGS = {
    'x': x,
    'y': y_2d,
    'mode': 'nearest'
}
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
latex_table Ran
ran; our contract draft was wrong, not the code
cvlab-kaist/locotrack/locotrack/evaluation_datasets.py
pointer only (licence: Apache-2.0); the stored copy get_code serves, from chopardda/GMA, is pointer only (licence: NONE) · get_code("17093d41c0b95c30")
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. A run is recorded.

posenc Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
cvlab-kaist/locotrack/locotrack/locotrack_model.py
code served (permissive licence) · get_code("e392edbfaa7809af")
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 2, dtype family float, shape (2, 28) 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

# Create a small test array in the valid range [-pi, pi]
x = jnp.array([
    [-3.0, -1.5, 0.0, 1.5],
    [0.5, -0.5, 3.0, -3.0]
], dtype=jnp.float32)

ARGS = {
    "x": x,
    "min_deg": 0,
    "max_deg": 3,
    "legacy_posenc_order": False,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
jax — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type ArrayImpl, ndim 2, dtype family float, shape (2, 28) — shape (2, 28) 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
prob_loss Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
cvlab-kaist/locotrack/locotrack/utils/model_utils.py
code served (permissive licence) · get_code("2db92844cacf356b")
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 numpy as np

# Set seed for reproducibility
np.random.seed(0)

# Define tiny test data
batch_size = 2
height = 3
width = 4
point_dim = 2

# tracks: shape (batch, height, width, 2) - tracked point coordinates
tracks = jnp.array(
    np.random.randn(batch_size, height, width, point_dim).astype(np.float32)
)

# expd: shape (batch, height, width) - expected/predicted values (logits or probabilities)
expd = jnp.array(
    np.random.randn(batch_size, height, width).astype(np.float32)
)

# target_points: shape (batch, height, width, 2) - target point coordinates
target_points = jnp.array(
    np.random.randn(batch_size, height, width, point_dim).astype(np.float32)
)

# occluded: shape (batch, height, width) - binary mask for occluded points (0 or 1)
occluded = jnp.array(
    np.random.randint(0, 2, size=(batch_size, height, width)).astype(np.float32)
)

# expected_dist_thresh: float threshold for distance
expected_dist_thresh = 8.0

ARGS = {
    "tracks": tracks,
    "expd": expd,
    "target_points": target_points,
    "occluded": occluded,
    "expected_dist_thresh": expected_dist_thresh,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
chex, jax, numpy, optax — 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
conv_block Not yet run
unverified (harvested, no recorded run)
cvlab-kaist/locotrack/locotrack/models/cmdtop.py
code served (permissive licence) · get_code("a8329e1fb8a56308")
How it ran

no invocation recorded.

huber_loss Not yet run
unverified (harvested, no recorded run)
cvlab-kaist/locotrack/locotrack_pytorch/model_utils.py
code served (permissive licence) · get_code("08fd9996236d683e")
How it ran

no invocation recorded.

huber_loss Not yet run
unverified (harvested, no recorded run)
cvlab-kaist/locotrack/locotrack/utils/model_utils.py
code served (permissive licence) · get_code("acafb736be892385")
How it ran

no invocation recorded.

prob_loss Not yet run
unverified (harvested, no recorded run)
cvlab-kaist/locotrack/locotrack_pytorch/model_utils.py
code served (permissive licence) · get_code("d71c81adf1b0a8dc")
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 (1 with at least one paper dated earlier, 0 the same month, 2 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
add_weight_decay 1 other paper
1 later
2411.09821 (2024-11)
Towards Scalable Newborn Screening: Automated General Movement Assessment in Uncontrolled Settings
latex_table 1 other paper
1 later
2411.09821 (2024-11)
Towards Scalable Newborn Screening: Automated General Movement Assessment in Uncontrolled Settings
posenc 1 other paper
1 dated earlier
2103.14024 (2021-03)
PlenOctrees for Real-time Rendering of Neural Radiance Fields

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

Fig. 1: Evaluating LocoTrack against state-of-the-art methods. We compare our LocoTrack against other SOTA methods [12,24] in terms of model size (circle size), position accuracy (y-axis), and throughput (xaxis). LocoTrack shows exceptionally high precision and efficiency.

For agents

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

get_harvested_code_for_paper("2407.15420")
get_code_for_paper("2407.15420")
have("2407.15420")

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

curl https://syntology.ai/api/ran/2407.15420.json

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

[![Syntology run record](https://syntology.ai/api/ran/2407.15420.svg)](https://syntology.ai/paper/2407.15420)

Connect an agent — have() is free.