SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2501.13667 · ICCV · 2025

MPG-SAM 2: Adapting SAM 2 with Mask Priors and Global Context for Referring Video Object Segmentation

Qian Zhang, Lefei Zhang, Meng Lan, Fu Rong

arXiv · PDF · Open in the Atlas

Code that ran

10 of 16 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 2 have no contract verdict; 6 failures of our instrument; 6 no recorded run), measured 2026-11-22; 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/2501.13667.json · badge: /api/ran/2501.13667.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
rongfu-dsb/MPG-SAM2 canonical (classifier label)
classifier basis not recorded on the link
10 ran · 6 no recorded run
FunctionStatusWhere it lives
compute_axial_cis Ran
ran, violated the contract we drafted
rongfu-dsb/MPG-SAM2/models/sam2/modeling/position_encoding.py
code served (permissive licence) · get_code("7ceaad42da42edd5")
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 (12, 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

torch.manual_seed(0)

ARGS = {
    "dim": 8,
    "end_x": 4,
    "end_y": 3,
    "theta": 10000.0,
}
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 (12, 4) — shape (12, 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
get_1d_sine_pe Ran
ran; our contract draft was wrong, not the code
rongfu-dsb/MPG-SAM2/models/sam2/modeling/sam2_utils.py
code served (permissive licence) · get_code("a6c3974188a5ffb0")
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 (2, 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

torch.manual_seed(0)

# pos_inds: a 1D tensor of position indices
pos_inds = torch.arange(2, dtype=torch.int64)

# dim: embedding dimension (should be even for sine/cosine split)
dim = 4

# temperature: temperature parameter for positional encoding (default 10000)
temperature = 10000

ARGS = {
    "pos_inds": pos_inds,
    "dim": dim,
    "temperature": temperature,
}
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 (2, 4) — shape (2, 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
get_activation_fn Ran
ran; our contract draft was wrong, not the code
rongfu-dsb/MPG-SAM2/models/sam2/modeling/sam2_utils.py
pointer only (licence: Apache-2.0); the stored copy get_code serves, from CGCL-codes/UAP-SAM2, is pointer only (licence: NONE) · get_code("f2e72d81f39d65cd")
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 CGCL-codes/UAP-SAM2, 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.

init_t_xy Ran
ran, honoured the contract we drafted
rongfu-dsb/MPG-SAM2/models/sam2/modeling/position_encoding.py
code served (permissive licence) · get_code("bb060d29525f90eb")
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 2, 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)

ARGS = {
    'end_x': 4,
    'end_y': 3,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type tuple, ndim 2, 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
is_box_near_crop_edge Ran
ran; our contract draft was wrong, not the code
rongfu-dsb/MPG-SAM2/models/sam2/utils/amg.py
code served (permissive licence) · get_code("3aac722b6505ee55")
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 (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 small batch of bounding boxes in xyxy format
# Shape: (2, 4) - 2 boxes with 4 coordinates each
boxes = torch.tensor([
    [10.0, 20.0, 50.0, 60.0],
    [100.0, 150.0, 200.0, 250.0]
], dtype=torch.float32)

# Crop box: [x0, y0, x1, y1]
crop_box = [5, 15, 300, 300]

# Original image box: [x0, y0, x1, y1]
orig_box = [0, 0, 512, 512]

# Tolerance parameter for nearness check
atol = 20.0

ARGS = {
    "boxes": boxes,
    "crop_box": crop_box,
    "orig_box": orig_box,
    "atol": atol
}
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 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
load_video_frames Ran
ran on a synthesized input
rongfu-dsb/MPG-SAM2/models/sam2/utils/misc.py
code served (permissive licence) · get_code("5cc4935655a7476a")
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. The code mentions cuda, and the harness aliases cuda to the CPU for every sample; whether this run needed that is 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 os
import tempfile
import torch
import numpy as np
from PIL import Image

torch.manual_seed(0)

# Create a temporary directory with JPEG frames
temp_dir = tempfile.mkdtemp()

# Create 3 simple JPEG frames
for i in range(3):
    # Create a simple RGB image
    img_array = np.random.randint(0, 256, (64, 64, 3), dtype=np.uint8)
    img = Image.fromarray(img_array, 'RGB')
    img.save(os.path.join(temp_dir, f'{i:05d}.jpg'))

ARGS = {
    'video_path': temp_dir,
    'image_size': 32,
    'offload_video_to_cpu': True,
    'img_mean': (0.485, 0.456, 0.406),
    'img_std': (0.229, 0.224, 0.225),
    'async_loading_frames': False,
    'compute_device': torch.device('cpu'),
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
PIL, numpy, torch, tqdm — 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
mask_to_box Ran
ran on a synthesized input
rongfu-dsb/MPG-SAM2/models/sam2/utils/misc.py
code served (permissive licence) · get_code("b91b1a2983dbf738")
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 3, dtype family int, shape (2, 1, 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

torch.manual_seed(0)

# Create a small mask tensor [B, 1, H, W]
# B=2 (batch size), 1 (channel), H=4 (height), W=3 (width)
masks = torch.zeros((2, 1, 4, 3), dtype=torch.bool)

# Add some True values to create actual masked regions
masks[0, 0, 1:3, 1:2] = True  # First batch has a mask region
masks[1, 0, 0:2, 0:3] = True  # Second batch has a different mask region

ARGS = {
    "masks": masks
}
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 3, dtype family int, shape (2, 1, 4) — shape (2, 1, 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
mask_to_rle_pytorch Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
rongfu-dsb/MPG-SAM2/models/sam2/utils/amg.py
code served (permissive licence) · get_code("bfa3b8111ef40247")
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 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 torch

torch.manual_seed(0)

# Create a small binary mask tensor with shape (batch, height, width)
# batch=2, height=4, width=3
tensor = torch.randint(0, 2, (2, 4, 3), dtype=torch.uint8)

ARGS = {
    'tensor': tensor
}
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, ndim 1, 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
reshape_for_broadcast Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
rongfu-dsb/MPG-SAM2/models/sam2/modeling/position_encoding.py
code served (permissive licence) · get_code("f8f09f78b474ee7c")
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 3, dtype family float, shape (1, 3, 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

torch.manual_seed(0)

# The function expects:
# - freqs_cis: a 2D tensor with shape (x.shape[-2], x.shape[-1])
# - x: a tensor with ndim >= 2

# Let's create x with shape (batch, seq_len, hidden_dim) = (2, 3, 4)
# This means x.shape[-2] = 3 and x.shape[-1] = 4
x = torch.randn(2, 3, 4)

# freqs_cis should have shape (3, 4) to match (x.shape[-2], x.shape[-1])
freqs_cis = torch.randn(3, 4)

ARGS = {
    "freqs_cis": freqs_cis,
    "x": x,
}
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 3, dtype family float, shape (1, 3, 4) — shape (1, 3, 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
select_closest_cond_frames Ran
ran; our contract draft was wrong, not the code
rongfu-dsb/MPG-SAM2/models/sam2/modeling/sam2_utils.py
code served (permissive licence) · get_code("4ee20f090e58c425")
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 torch
import numpy as np

torch.manual_seed(0)

# Construct cond_frame_outputs as a dictionary mapping frame indices to output values
# The function uses cond_frame_outputs as a dict with integer keys (frame indices)
# and arbitrary values. We'll use tensors as stand-in output values.
cond_frame_outputs = {
    0: torch.randn(2, 4),
    5: torch.randn(2, 4),
    10: torch.randn(2, 4),
    15: torch.randn(2, 4),
    20: torch.randn(2, 4),
    25: torch.randn(2, 4),
}

# frame_idx is the current frame index we're selecting conditioning frames for
frame_idx = 12

# max_cond_frame_num is the maximum number of conditioning frames to select
# Use a value that requires selection logic (not -1 and not >= len(cond_frame_outputs))
max_cond_frame_num = 3

ARGS = {
    "frame_idx": frame_idx,
    "cond_frame_outputs": cond_frame_outputs,
    "max_cond_frame_num": max_cond_frame_num,
}
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 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
box_xyxy_to_xywh Not yet run
unverified (harvested, no recorded run)
rongfu-dsb/MPG-SAM2/models/sam2/utils/amg.py
code served (permissive licence) · get_code("ca78fd1ad096a766")
How it ran

no invocation recorded.

build_sam2 Not yet run
unverified (harvested, no recorded run)
rongfu-dsb/MPG-SAM2/models/sam2/build_sam.py
code served (permissive licence) · get_code("df0caf7e54b2c61e")
How it ran

no invocation recorded.

build_sam2_hf Not yet run
unverified (harvested, no recorded run)
rongfu-dsb/MPG-SAM2/models/sam2/build_sam.py
code served (permissive licence) · get_code("c83fb75d631af894")
How it ran

no invocation recorded.

build_sam2_video_predictor Not yet run
unverified (harvested, no recorded run)
rongfu-dsb/MPG-SAM2/models/sam2/build_sam.py
code served (permissive licence) · get_code("b6bf96dd6b61e154")
How it ran

no invocation recorded.

get_connected_components Not yet run
unverified (harvested, no recorded run)
rongfu-dsb/MPG-SAM2/models/sam2/utils/misc.py
code served (permissive licence) · get_code("baff5f8846f48223")
How it ran

no invocation recorded.

parse_args Not yet run
unverified (harvested, no recorded run)
rongfu-dsb/MPG-SAM2/merge_lora_weights_and_save_hf_model.py
code served (permissive licence) · get_code("77c6e8c3acbaea40")
How it ran

no invocation recorded.

Also present in other papers' code

Of 16 samples lifted from this paper's linked repositories, 16 carry a function identity and 0 do not. Of the 16 distinct functions with an identity, 11 are also present in other papers whose code we hold (11 with at least one paper dated earlier, 3 the same month, 10 later) and 5 are only here among the 52,977 papers whose code Syntology holds (as of 2026-09-26).

Only here means: among the 52,977 papers whose code Syntology holds (51,887 with lifted samples, 1,090 through their assemblies only), as of 2026-09-26, no other paper's lifted samples or current assembly bricks contain an identical function body after normalisation.

FunctionAlso present inEarliest-datedOthers named
compute_axial_cis 13 other papers
9 dated earlier, 1 the same month, 3 later
2302.01872 (2023-02)
MOSE: A New Dataset for Video Object Segmentation in Complex Scenes
2306.01567 (2023-06)
Segment Anything in High Quality
2408.00874 (2024-08)
Medical SAM 2: Segment medical images as video via Segment Anything Model 2
and 10 more not named here
init_t_xy 13 other papers
9 dated earlier, 1 the same month, 3 later
2302.01872 (2023-02)
MOSE: A New Dataset for Video Object Segmentation in Complex Scenes
2306.01567 (2023-06)
Segment Anything in High Quality
2408.00874 (2024-08)
Medical SAM 2: Segment medical images as video via Segment Anything Model 2
and 10 more not named here
reshape_for_broadcast 13 other papers
9 dated earlier, 1 the same month, 3 later
2302.01872 (2023-02)
MOSE: A New Dataset for Video Object Segmentation in Complex Scenes
2306.01567 (2023-06)
Segment Anything in High Quality
2408.00874 (2024-08)
Medical SAM 2: Segment medical images as video via Segment Anything Model 2
and 10 more not named here
build_sam2 10 other papers
5 dated earlier, 5 later
2302.01872 (2023-02)
MOSE: A New Dataset for Video Object Segmentation in Complex Scenes
2306.01567 (2023-06)
Segment Anything in High Quality
2408.07931 (2024-08)
Surgical SAM 2: Real-time Segment Anything in Surgical Video by Efficient Frame Pruning
and 7 more not named here
get_activation_fn 10 other papers
6 dated earlier, 4 later
2404.07445 (2024-04)
Multi-view Aggregation Network for Dichotomous Image Segmentation
2407.10749 (2024-07)
SEED: A Simple and Effective 3D DETR in Point Clouds
2408.00874 (2024-08)
Medical SAM 2: Segment medical images as video via Segment Anything Model 2
and 7 more not named here
get_1d_sine_pe 9 other papers
5 dated earlier, 4 later · 1 only as an assembly brick
2408.00714 (2024-08)
SAM 2: Segment Anything in Images and Videos
2408.00874 (2024-08)
Medical SAM 2: Segment medical images as video via Segment Anything Model 2
2408.07931 (2024-08)
Surgical SAM 2: Real-time Segment Anything in Surgical Video by Efficient Frame Pruning
and 6 more not named here
select_closest_cond_frames 9 other papers
5 dated earlier, 4 later · 1 only as an assembly brick
2408.00714 (2024-08)
SAM 2: Segment Anything in Images and Videos
2408.00874 (2024-08)
Medical SAM 2: Segment medical images as video via Segment Anything Model 2
2408.07931 (2024-08)
Surgical SAM 2: Real-time Segment Anything in Surgical Video by Efficient Frame Pruning
and 6 more not named here
box_xyxy_to_xywh 7 other papers
6 dated earlier, 1 later
2305.10724 (2023-05)
Segment Any Anomaly without Training via Hybrid Prompt Regularization
2308.07624 (2023-08)
Self-Prompting Large Vision Models for Few-Shot Medical Image Segmentation
2309.06824 (2023-09)
Beyond Adapting SAM: Towards End-to-End Ultrasound Image Segmentation via Auto Prompting
and 4 more not named here
is_box_near_crop_edge 6 other papers
5 dated earlier, 1 later
2305.10724 (2023-05)
Segment Any Anomaly without Training via Hybrid Prompt Regularization
2308.07624 (2023-08)
Self-Prompting Large Vision Models for Few-Shot Medical Image Segmentation
2309.06824 (2023-09)
Beyond Adapting SAM: Towards End-to-End Ultrasound Image Segmentation via Auto Prompting
and 3 more not named here
mask_to_rle_pytorch 6 other papers
5 dated earlier, 1 later
2305.10724 (2023-05)
Segment Any Anomaly without Training via Hybrid Prompt Regularization
2308.07624 (2023-08)
Self-Prompting Large Vision Models for Few-Shot Medical Image Segmentation
2309.06824 (2023-09)
Beyond Adapting SAM: Towards End-to-End Ultrasound Image Segmentation via Auto Prompting
and 3 more not named here
build_sam2_hf 2 other papers
2 dated earlier
2408.07931 (2024-08)
Surgical SAM 2: Real-time Segment Anything in Surgical Video by Efficient Frame Pruning
2408.16768 (2024-08)
SAM2Point: Segment Any 3D as Videos in Zero-shot and Promptable Manners

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

Referring video object segmentation (RVOS) aims to segment objects in a video according to textual descriptions, which requires the integration of multimodal information and temporal dynamics perception. The Segment Anything Model 2 (SAM 2) has shown great effectiveness across various video segmentation tasks. However, its application to offline RVOS is challenged by the translation of the text into effective prompts and a lack of global context awareness. In this paper, we propose a novel RVOS framework, termed MPG-SAM 2, to address these challenges. Specifically, MPG-SAM 2 employs a multimodal encoder to jointly encode video and textual features, generating semantically aligned video and text embeddings along with multimodal class tokens. A mask prior generator is devised to utilize the video embeddings and class tokens to create pseudo masks of target objects and global context. These masks are fed into the prompt encoder as dense prompts, along with multimodal class tokens as sparse prompts to generate accurate prompts for SAM 2. To provide the online SAM 2 with a global view, we propose a hierarchical global-historical aggregator, which allows SAM 2 to aggregate global and historical information of target objects at both pixel and object levels, enhancing the target representation and temporal consistency. Extensive experiments on several RVOS benchmarks demonstrate the superiority of MPG-SAM 2 and the effectiveness of the proposed modules. The code is available at https://github.com/rongfu-dsb/MPG-SAM2.

For agents

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

get_harvested_code_for_paper("2501.13667")
get_code_for_paper("2501.13667")
have("2501.13667")

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

curl https://syntology.ai/api/ran/2501.13667.json

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

[![Syntology run record](https://syntology.ai/api/ran/2501.13667.svg)](https://syntology.ai/paper/2501.13667)

Connect an agent — have() is free.