SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2406.04659 · CVPR · 2024

LocLLM: Exploiting Generalizable Human Keypoint Localization via Large Language Model

Shiliang Zhang, Dongkai Wang, Shiyu Xuan

arXiv · PDF · Open in the Atlas

Code that ran

11 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, 1 violated it and 0 have no contract verdict; 10 failures of our instrument; 1 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/2406.04659.json · badge: /api/ran/2406.04659.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
kennethwdk/LocLLM canonical (classifier label)
classifier basis: census
11 ran · 1 no recorded run
FunctionStatusWhere it lives
add_residual Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
kennethwdk/LocLLM/models/dino_layers/block.py
code served (permissive licence) · get_code("5d16d4d4fc573ac7")
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)

# Build tiny tensors for the function arguments
# x: input tensor, shape (batch=2, features=4)
x = torch.randn(2, 4, dtype=torch.float32)

# brange: indices for where to add residual, shape should be compatible with batch dimension
brange = torch.tensor([0, 1], dtype=torch.long)

# residual: tensor to add, same shape as x
residual = torch.randn(2, 4, dtype=torch.float32)

# residual_scale_factor: scalar multiplier
residual_scale_factor = 0.5

# scaling_vector: optional vector for scaled operations, set to None for first branch
scaling_vector = None

ARGS = {
    "x": x,
    "brange": brange,
    "residual": residual,
    "residual_scale_factor": residual_scale_factor,
    "scaling_vector": scaling_vector,
}
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
affine_transform Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
kennethwdk/LocLLM/datasets/h36m.py
pointer only (licence: MIT); the stored copy get_code serves, from kmax2001/EvSharp2Blur, is pointer only (licence: NONE) · get_code("83abf1ce90db7100")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (MIT), but get_code serves the stored copy, first harvested from kmax2001/EvSharp2Blur, 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.

disabled_train Ran
ran, violated the contract we drafted
kennethwdk/LocLLM/models/locllm.py
code served (permissive licence) · get_code("4cb732f513d69dfd")
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) and returned SimpleModel. 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

torch.manual_seed(0)

class SimpleModel(nn.Module):
    def __init__(self):
        super().__init__()
        self.linear = nn.Linear(4, 3)
    
    def forward(self, x):
        return self.linear(x)

model = SimpleModel()

ARGS = {
    "self": model,
    "mode": True
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type SimpleModel — a SimpleModel output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a SimpleModel 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
drop_add_residual_stochastic_depth Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
kennethwdk/LocLLM/models/dino_layers/block.py
code served (permissive licence) · get_code("85f7ffc01945bb72")
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 (2, 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
from torch import nn, Tensor
from typing import Callable

torch.manual_seed(0)

# Create a simple residual function that takes a tensor and returns a tensor of same shape
class SimpleResidualFunc(nn.Module):
    def __init__(self, dim: int):
        super().__init__()
        self.linear = nn.Linear(dim, dim)
    
    def forward(self, x: Tensor) -> Tensor:
        # x will have shape (sample_subset_size, n, d)
        # We need to return same shape
        b, n, d = x.shape
        x_flat = x.flatten(1)  # (b, n*d)
        out = self.linear(x_flat)  # (b, n*d)
        return out.view(b, n, d)

# Create input tensor with shape (batch=2, n=3, d=4)
x = torch.randn(2, 3, 4)

# Create a residual function
residual_func = SimpleResidualFunc(dim=3*4)

# Define the sample_drop_ratio parameter
sample_drop_ratio = 0.1

ARGS = {
    "x": x,
    "residual_func": residual_func,
    "sample_drop_ratio": sample_drop_ratio
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Tensor, ndim 3, dtype family float, shape (2, 3, 4) — shape (2, 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
flip_joints_3d Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
kennethwdk/LocLLM/datasets/h36m.py
code served (permissive licence) · get_code("88428d9a8f104e23")
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 float, shape (5, 3, 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 as np

# Create a small valid input for flip_joints_3d
# joints_3d should be shape (num_joints, 3, 2)
num_joints = 5
joints_3d = np.random.rand(num_joints, 3, 2).astype(np.float32)

# width is an integer representing image width
width = 640

# joint_pairs is a list of tuples representing joint index pairs to swap
joint_pairs = [(0, 1), (2, 3)]

ARGS = {
    "joints_3d": joints_3d,
    "width": width,
    "joint_pairs": joint_pairs,
}
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 3, dtype family float, shape (5, 3, 2) — shape (5, 3, 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_3rd_point Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
kennethwdk/LocLLM/datasets/h36m.py
code served (permissive licence) · get_code("f8c01e13c130a2a9")
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 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 numpy as np

# Construct valid arguments for get_3rd_point
# The function takes two parameters: a and b
# Both should be numpy arrays (based on the function's use of subtraction and indexing)
# The function expects 2D points (uses direct[0] and direct[1])

a = np.array([3.0, 4.0], dtype=np.float32)
b = np.array([1.0, 2.0], dtype=np.float32)

ARGS = {
    "a": a,
    "b": b
}
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 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
get_affine_transform Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
kennethwdk/LocLLM/datasets/coco.py
pointer only (licence: MIT); the stored copy get_code serves, from kmax2001/EvSharp2Blur, is pointer only (licence: NONE) · get_code("70696932e5ece0fb")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (MIT), but get_code serves the stored copy, first harvested from kmax2001/EvSharp2Blur, 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.

get_branges_scales Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
kennethwdk/LocLLM/models/dino_layers/block.py
code served (permissive licence) · get_code("5be3610fa1fee19e")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned tuple. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch

torch.manual_seed(0)

# Construct a small tensor with shape (batch=2, sequence=4, dim=3)
x = torch.randn(2, 4, 3)

# Define arguments
ARGS = {
    "x": x,
    "sample_drop_ratio": 0.0
}
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 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
lora_state_dict Ran
ran; our contract draft was wrong, not the code
kennethwdk/LocLLM/models/lora.py
code served (permissive licence) · get_code("e7af7e6c543635d7")
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 torch
import torch.nn as nn

torch.manual_seed(0)

class LoRALayer(nn.Module):
    """A simple layer with LoRA matrices A and B"""
    def __init__(self):
        super().__init__()
        self.lora_A = nn.Parameter(torch.randn(4, 3))
        self.lora_B = nn.Parameter(torch.randn(3, 4))
        self.bias = nn.Parameter(torch.randn(4))
    
    def forward(self, x):
        return x

class ModelWithLoRA(nn.Module):
    """Model containing LoRA layers"""
    def __init__(self):
        super().__init__()
        self.lora_layer = LoRALayer()
        self.regular_layer = nn.Linear(4, 2)
    
    def forward(self, x):
        return x

model = ModelWithLoRA()

ARGS = {
    'model': model,
    'bias': 'none'
}
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 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
named_apply Ran
ran; our contract draft was wrong, not the code
kennethwdk/LocLLM/models/dino.py
code served (permissive licence) · get_code("e7fcb6d9ac9deaf4")
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) and returned SimpleModel. 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

torch.manual_seed(0)

# Create a simple nested module structure
class SimpleModel(nn.Module):
    def __init__(self):
        super().__init__()
        self.layer1 = nn.Linear(4, 3)
        self.layer2 = nn.Sequential(
            nn.Linear(3, 2),
            nn.ReLU()
        )
    
    def forward(self, x):
        x = self.layer1(x)
        x = self.layer2(x)
        return x

# Create a simple callback function that tracks module names
collected_modules = []

def callback_fn(module, name):
    collected_modules.append((name, type(module).__name__))

# Instantiate the module
module = SimpleModel()

# Define ARGS dictionary with all parameters
ARGS = {
    'fn': callback_fn,
    'module': module,
    'name': '',
    'depth_first': True,
    'include_root': False
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type SimpleModel — a SimpleModel output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a SimpleModel 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
transform_preds Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
kennethwdk/LocLLM/datasets/coco.py
code served (permissive licence) · get_code("74b80552a8a3db3c")
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, 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

ARGS = {
    'coords': np.array([[10.5, 20.3, 5.0], [15.2, 25.1, 8.0]], dtype=np.float32),
    'center': np.array([64.0, 64.0], dtype=np.float32),
    'scale': np.array([0.5, 0.5], dtype=np.float32),
    'output_size': (128, 128)
}
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 2, dtype family float, shape (2, 3) — shape (2, 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
fliplr_joints Not yet run
unverified (harvested, no recorded run)
kennethwdk/LocLLM/datasets/coco.py
code served (permissive licence) · get_code("d49a474bc36c5234")
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, 11 are also present in other papers whose code we hold (10 with at least one paper dated earlier, 6 the same month, 9 later) and 1 are only here among the 52,977 papers whose code Syntology holds (as of 2026-09-26).

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

FunctionAlso present inEarliest-datedOthers named
disabled_train 173 other papers
117 dated earlier, 9 the same month, 39 later, 8 same year, order not determinable · 5 only as an assembly brick
2012.09841 (2020-12)
Taming Transformers for High-Resolution Image Synthesis
2102.05379 (2021-02)
Argmax Flows and Multinomial Diffusion: Learning Categorical Distributions
2104.07652 (2021-04)
Geometry-Free View Synthesis: Transformers and no 3D Priors
and 170 more not named here
named_apply 26 other papers
8 dated earlier, 1 the same month, 15 later, 2 undated
2010.11929 (2020-10)
An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale
2204.07683 (2022-04)
Safe Self-Refinement for Transformer-based Domain Adaptation
2304.07193 (2023-04)
DINOv2: Learning Robust Visual Features without Supervision
and 23 more not named here
drop_add_residual_stochastic_depth 21 other papers
5 dated earlier, 2 the same month, 12 later, 2 undated
2010.11929 (2020-10)
An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale
2304.07193 (2023-04)
DINOv2: Learning Robust Visual Features without Supervision
2312.02366 (2023-12)
Evaluating General Purpose Vision Foundation Models for Medical Image Analysis: An Experimental Study of DINOv2 on Radiology Benchmarks
and 18 more not named here
add_residual 17 other papers
4 dated earlier, 2 the same month, 10 later, 1 undated
2304.07193 (2023-04)
DINOv2: Learning Robust Visual Features without Supervision
2312.02366 (2023-12)
Evaluating General Purpose Vision Foundation Models for Medical Image Analysis: An Experimental Study of DINOv2 on Radiology Benchmarks
2402.03138 (2024-02)
Just Cluster It: An Approach for Exploration in High-Dimensions using Clustering and Pre-Trained Representations
and 14 more not named here
get_branges_scales 17 other papers
4 dated earlier, 2 the same month, 10 later, 1 undated
2304.07193 (2023-04)
DINOv2: Learning Robust Visual Features without Supervision
2312.02366 (2023-12)
Evaluating General Purpose Vision Foundation Models for Medical Image Analysis: An Experimental Study of DINOv2 on Radiology Benchmarks
2402.03138 (2024-02)
Just Cluster It: An Approach for Exploration in High-Dimensions using Clustering and Pre-Trained Representations
and 14 more not named here
lora_state_dict 13 other papers
6 dated earlier, 1 the same month, 5 later, 1 same year, order not determinable
2202.00535 (2022-02)
Novelty Controlled Paraphrase Generation with Retrieval Augmented Conditional Prompt Tuning
2202.07962 (2022-02)
Revisiting Parameter-Efficient Tuning: Are We Really There Yet?
2211.09790 (2022-11)
ConStruct-VL: Data-Free Continual Structured VL Concepts Learning
and 10 more not named here
get_3rd_point 5 other papers
4 dated earlier, 1 later
1906.07155 (2019-06)
MMDetection: Open MMLab Detection Toolbox and Benchmark
1908.07919 (2019-08)
Deep High-Resolution Representation Learning for Visual Recognition
2105.04154 (2021-05)
Unsupervised Human Pose Estimation through Transforming Shape Templates
and 2 more not named here
transform_preds 4 other papers
4 dated earlier
1906.07155 (2019-06)
MMDetection: Open MMLab Detection Toolbox and Benchmark
2006.11275 (2020-06)
Center-based 3D Object Detection and Tracking
2007.09548 (2020-07)
Kinematic 3D Object Detection in Monocular Video
and 1 more not named here
affine_transform 2 other papers
1 dated earlier, 1 later
2007.09548 (2020-07)
Kinematic 3D Object Detection in Monocular Video
2507.22438 (2025-07)
From Sharp to Blur: Unsupervised Domain Adaptation for 2D Human Pose Estimation Under Extreme Motion Blur Using Event Cameras
get_affine_transform 2 other papers
2 later
2507.22438 (2025-07)
From Sharp to Blur: Unsupervised Domain Adaptation for 2D Human Pose Estimation Under Extreme Motion Blur Using Event Cameras
2602.03157 (2026-02)
Human-in-the-loop adaptation in group activity feature learning for team sports video retrieval
fliplr_joints 1 other paper
1 dated earlier
2208.00453 (2022-08)
One-Shot Medical Landmark Localization by Edge-Guided Transform and Noisy Landmark Refinement

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 capacity of existing human keypoint localization models is limited by keypoint priors provided by the training data. To alleviate this restriction and pursue more general model, this work studies keypoint localization from a different perspective by reasoning locations based on keypiont clues in text descriptions. We propose LocLLM, the first Large-Language Model (LLM) based keypoint localization model that takes images and text instructions as inputs and outputs the desired keypoint coordinates. LocLLM leverages the strong reasoning capability of LLM and clues of keypoint type, location, and relationship in textual descriptions for keypoint localization. To effectively tune Lo-cLLM, we construct localization-based instruction conversations to connect keypoint description with corresponding coordinates in input image, and fine-tune the whole model in a parameter-efficient training pipeline. LocLLM shows remarkable performance on standard 2D/3D keypoint localization benchmarks. Moreover, incorporating language clues into the localization makes LocLLM show superior flexibility and generalizable capability in cross dataset keypoint localization, and even detecting novel type of keypoints unseen during training † .

For agents

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

get_harvested_code_for_paper("2406.04659")
get_code_for_paper("2406.04659")
have("2406.04659")

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

curl https://syntology.ai/api/ran/2406.04659.json

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

[![Syntology run record](https://syntology.ai/api/ran/2406.04659.svg)](https://syntology.ai/paper/2406.04659)

Connect an agent — have() is free.