SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2211.11682 · 2022

PointCLIP V2: Prompting CLIP and GPT for Powerful 3D Open-world Learning

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; 0 honoured the contract we drafted, 1 violated it and 3 have no contract verdict; 4 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/2211.11682.json · badge: /api/ran/2211.11682.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
yangyangyang127/pointclip_v2 canonical (classifier label)
classifier basis not recorded on the link
8 ran · 4 no recorded run
FunctionStatusWhere it lives
basic_clean Ran
ran; our contract draft was wrong, not the code
yangyangyang127/pointclip_v2/zeroshot_cls/clip/simple_tokenizer.py
pointer only (licence: MIT); the stored copy get_code serves, from ltpwy/MSCI, is pointer only (licence: NONE) · get_code("98f385d847636a3e")
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 ltpwy/MSCI, 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.

cal_loss Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
yangyangyang127/pointclip_v2/zeroshot_seg/util.py
code served (permissive licence) · get_code("c6c2758c0c2fa756")
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 Tensor, ndim 0, dtype family float, shape () as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
import torch.nn.functional as F

torch.manual_seed(0)

# Create small tensors for testing
batch_size = 2
num_classes = 4

# pred: logits tensor of shape (batch_size, num_classes)
pred = torch.randn(batch_size, num_classes)

# gold: target class indices of shape (batch_size,) or can be reshaped
gold = torch.tensor([0, 2])

# smoothing: boolean flag
smoothing = True

ARGS = {
    "pred": pred,
    "gold": gold,
    "smoothing": smoothing
}
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 Tensor, ndim 0, dtype family float, shape () — shape () as a later run recorded it
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
calculate_shape_IoU Ran
ran on a synthesized input
yangyangyang127/pointclip_v2/zeroshot_seg/util.py
code served (permissive licence) · get_code("d51a6e40c0aa5d4f")
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 numpy as np
import torch

# Analyze the function to understand what it needs:
# - pred_np: numpy array, indexed as pred_np[shape_idx], compared with == operator
# - seg_np: numpy array, same shape as pred_np
# - label: tensor or array that can be squeezed, indexed by shape_idx
# - class_choice: boolean or None (used in if not class_choice)
# - eva: boolean (default False)

# The function iterates over seg_np.shape[0] (batch dimension)
# pred_np and seg_np are compared with part indices (integers 0-6)

torch.manual_seed(0)

# Create small numpy arrays for pred_np and seg_np
# Each shape has a flat array of part predictions/ground truth
# Batch size 2, with part indices ranging 0-5
batch_size = 2
num_points = 50

pred_np = np.random.randint(0, 6, size=(batch_size, num_points), dtype=np.int32)
seg_np = np.random.randint(0, 6, size=(batch_size, num_points), dtype=np.int32)

# Create label tensor - should have batch_size elements
# Values should be valid category indices (0-15 based on seg_num and index_start)
label = torch.tensor([[0], [5]], dtype=torch.long)  # batch_size x 1

# class_choice: False means use index_start lookup, True means use seg_num directly
class_choice = False

# eva: whether to return tuple or just list
eva = True

ARGS = {
    "pred_np": pred_np,
    "seg_np": seg_np,
    "label": label,
    "class_choice": class_choice,
    "eva": eva,
}
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 — 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
euler2mat Ran
ran, violated the contract we drafted
yangyangyang127/pointclip_v2/zeroshot_seg/realistic_projection.py
pointer only (licence: MIT); the stored copy get_code serves, from ivan-tang-3d/any2point, is pointer only (licence: NONE) · get_code("140122b970061273")
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 ivan-tang-3d/any2point, 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.

getGaussianKernel2D Ran
ran on a synthesized input
yangyangyang127/pointclip_v2/zeroshot_seg/realistic_projection.py
code served (permissive licence) · get_code("9b099a224dbca79d")
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 (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 numpy as np

ARGS = {
    "ksize": 5,
    "sigma": 1.0
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type ndarray, ndim 2, dtype family float, shape (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
get_pairs Ran
ran; our contract draft was wrong, not the code
yangyangyang127/pointclip_v2/zeroshot_cls/clip/simple_tokenizer.py
code served (permissive licence) · get_code("d919ae32e5e4e616")
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 set. 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)

# The function get_pairs takes a word parameter
# A word is represented as a tuple of symbols (variable-length strings)
# The function iterates through the word and creates pairs of consecutive symbols

word = ("hello", "world", "test")

ARGS = {
    "word": word
}
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 set — a set output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a set 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
points2grid Ran
ran on a synthesized input
yangyangyang127/pointclip_v2/zeroshot_seg/realistic_projection.py
code served (permissive licence) · get_code("706ac9d3b62c54e8")
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)

# Define global params and net for the function
params = {'vit_b16': {'maxpoolz': 5, 'maxpoolxy': 11, 'maxpoolpadz': 2, 'maxpoolpadxy': 5, 'convz': 5, 'convxy': 5, 'convsigmaxy': 1, 'convsigmaz': 2, 'convpadz': 2, 'convpadxy': 2, 'imgbias': 0.0, 'depth_bias': 0.3, 'obj_ratio': 0.7, 'bg_clr': 0.0, 'resolution': 224, 'depth': 112}}

net = 'vit_b16'

# Create sample points tensor
# Shape: [batch=2, pnum=4, coords=3]
# Points should be in some normalized range for the function to work
points = torch.randn(2, 4, 3) * 2.0  # Random 3D points

image_height = 224
image_width = 224
depth = params[net]['depth']
device = 'cpu'

ARGS = {
    'points': points,
    'image_height': image_height,
    'image_width': image_width,
    'depth': depth,
    'device': device
}
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
whitespace_clean Ran
ran; our contract draft was wrong, not the code
yangyangyang127/pointclip_v2/zeroshot_cls/clip/simple_tokenizer.py
pointer only (licence: MIT); the stored copy get_code serves, from ltpwy/MSCI, is pointer only (licence: NONE) · get_code("9542161e9640b858")
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 ltpwy/MSCI, 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.

build_model Not yet run
unverified (harvested, no recorded run)
yangyangyang127/pointclip_v2/zeroshot_cls/clip/model.py
pointer only (licence: MIT); the stored copy get_code serves, from jayparanjape/biastuning, is pointer only (licence: NONE) · get_code("69c02e4eb1c02a52")
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 jayparanjape/biastuning, 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.

build_model Not yet run
unverified (harvested, no recorded run)
yangyangyang127/pointclip_v2/zeroshot_seg/clip/model.py
code served (permissive licence) · get_code("e475fd539a310ed2")
How it ran

no invocation recorded.

load Not yet run
unverified (harvested, no recorded run)
yangyangyang127/pointclip_v2/zeroshot_cls/clip/clip.py
code served (permissive licence) · get_code("fbf8c0143d9c48e3")
How it ran

no invocation recorded.

load_data_partseg Not yet run
unverified (harvested, no recorded run)
yangyangyang127/pointclip_v2/zeroshot_seg/data.py
code served (permissive licence) · get_code("6ac515bebb5bb563")
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 11 distinct functions with an identity, 7 are also present in other papers whose code we hold (6 with at least one paper dated earlier, 6 the same month, 7 later) and 4 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
get_pairs 258 other papers
31 dated earlier, 3 the same month, 223 later, 1 undated · 4 only as an assembly brick
1802.06474 (2018-02)
A Closed-form Solution to Photorealistic Image Stylization
1803.01128 (2018-03)
Seq2Sick: Evaluating the Robustness of Sequence-to-Sequence Models with Adversarial Examples
1809.00125 (2018-09)
Simple Fusion: Return of the Language Model
and 255 more not named here
whitespace_clean 231 other papers
17 dated earlier, 3 the same month, 210 later, 1 undated · 4 only as an assembly brick
1901.07441 (2019-01)
PadChest: A large chest x-ray image dataset with multi-label annotated reports
2107.06383 (2021-07)
How Much Can CLIP Benefit Vision-and-Language Tasks?
2108.10904 (2021-08)
SimVLM: Simple Visual Language Model Pretraining with Weak Supervision
and 228 more not named here
basic_clean 218 other papers
17 dated earlier, 3 the same month, 197 later, 1 undated · 4 only as an assembly brick
1901.07441 (2019-01)
PadChest: A large chest x-ray image dataset with multi-label annotated reports
2107.06383 (2021-07)
How Much Can CLIP Benefit Vision-and-Language Tasks?
2108.10904 (2021-08)
SimVLM: Simple Visual Language Model Pretraining with Weak Supervision
and 215 more not named here
build_model 119 other papers
18 dated earlier, 3 the same month, 96 later, 2 undated
1809.11096 (2018-09)
Large Scale GAN Training for High Fidelity Natural Image Synthesis
1901.07441 (2019-01)
PadChest: A large chest x-ray image dataset with multi-label annotated reports
2104.14294 (2021-04)
Emerging Properties in Self-Supervised Vision Transformers
and 116 more not named here
load 88 other papers
5 dated earlier, 1 the same month, 81 later, 1 undated
2203.05557 (2022-03)
Conditional Prompt Learning for Vision-Language Models
2206.01986 (2022-06)
Delving into the Openness of CLIP
2207.12819 (2022-07)
S-Prompts Learning with Pre-trained Transformers: An Occam's Razor for Domain Incremental Learning
and 85 more not named here
cal_loss 7 other papers
5 dated earlier, 1 the same month, 1 later
1911.00195 (2019-11)
Rotation Invariant Point Cloud Classification: Where Local Geometry Meets Global Topology
2101.01461 (2021-01)
PointCutMix: Regularization Strategy for Point Cloud Classification
2201.12296 (2022-01)
Benchmarking Robustness of 3D Point Cloud Recognition Against Common Corruptions
and 4 more not named here
euler2mat 1 other paper
1 later
2303.16894 (2023-03)
ViewRefer: Grasp the Multi-view Knowledge for 3D Visual Grounding

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

Large-scale pre-trained models have shown promising open-world performance for both vision and language tasks. However, their transferred capacity on 3D point clouds is still limited and only constrained to the classification task. In this paper, we first collaborate CLIP and GPT to be a unified 3D open-world learner, named as PointCLIP V2, which fully unleashes their potential for zero-shot 3D classification, segmentation, and detection. To better align 3D data with the pre-trained language knowledge, PointCLIP V2 contains two key designs. For the visual end, we prompt CLIP via a shape projection module to generate more realistic depth maps, narrowing the domain gap between projected point clouds with natural images. For the textual end, we prompt the GPT model to generate 3D-specific text as the input of CLIP's textual encoder. Without any training in 3D domains, our approach significantly surpasses PointCLIP by +42.90%, +40.44%, and +28.75% accuracy on three datasets for zero-shot 3D classification. On top of that, V2 can be extended to few-shot 3D classification, zero-shot 3D part segmentation, and 3D object detection in a simple manner, demonstrating our generalization ability for unified 3D open-world learning.

For agents

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

get_harvested_code_for_paper("2211.11682")
get_code_for_paper("2211.11682")
have("2211.11682")

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

curl https://syntology.ai/api/ran/2211.11682.json

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

[![Syntology run record](https://syntology.ai/api/ran/2211.11682.svg)](https://syntology.ai/paper/2211.11682)

Connect an agent — have() is free.