SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2503.03751 · CVPR · 2025

GEN3C: 3D-Informed World-Consistent Video Generation with Precise Camera Control

Huan Ling, Jun Gao, Sanja Fidler, Yifan Lu, Xuanchi Ren, Tianchang Shen, Jiahui Huang, Alexander Keller, Thomas Müller, Merlin Nimier-David

arXiv · PDF · Open in the Atlas

Code that ran

2 of 2 samples (functions and classes) lifted from this paper's linked repositories executed on a synthesized input in Syntology's jail (0 of them constructed an object rather than computing a result; 0 honoured the contract we drafted, 0 violated it and 2 have no contract verdict; 0 failures of our instrument; 0 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/2503.03751.json · badge: /api/ran/2503.03751.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
nv-tlabs/GEN3C canonical (classifier label)
classifier basis not recorded on the link
2 ran · 0 no recorded run
FunctionStatusWhere it lives
resize_intrinsics Ran
ran on a synthesized input
nv-tlabs/GEN3C/cosmos_predict1/diffusion/inference/gen3c_persistent.py
code served (permissive licence) · get_code("25b0f02063e9d156")
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 (2, 3, 3) as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
import numpy as np

torch.manual_seed(0)

# The function expects:
# - intrinsics: np.ndarray or torch.Tensor with shape (batch, 3, 3) for camera intrinsics matrices
# - old_size: tuple of (height, width) before resizing
# - new_size: tuple of (height, width) after resizing
# - crop_size: optional tuple of (height, width) for cropping

# Create a batch of intrinsic matrices (3x3 camera matrices)
# Shape: (batch_size=2, 3, 3)
intrinsics_np = np.array([
    [[500.0, 0.0, 320.0],
     [0.0, 500.0, 240.0],
     [0.0, 0.0, 1.0]],
    [[600.0, 0.0, 320.0],
     [0.0, 600.0, 240.0],
     [0.0, 0.0, 1.0]]
])

old_size = (480, 640)  # height, width
new_size = (240, 320)  # height, width (downsampled)
crop_size = (200, 300)  # optional crop dimensions

ARGS = {
    'intrinsics': intrinsics_np,
    'old_size': old_size,
    'new_size': new_size,
    'crop_size': crop_size
}
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 ndarray, ndim 3, dtype family float, shape (2, 3, 3) — shape (2, 3, 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
scaled_dot_product_attention Ran
ran on a synthesized input
nv-tlabs/GEN3C/cosmos_predict1/autoregressive/modules/attention.py
code served (permissive licence) · get_code("abc31b2f673701c7")
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 4, dtype family float, shape (2, 3, 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
import math

torch.manual_seed(0)

# Tiny tensors for attention computation
# Shape: (batch_size, seq_len, num_heads, head_dim)
batch_size = 2
seq_len = 3
num_heads = 2
head_dim = 4

q = torch.randn(batch_size, num_heads, seq_len, head_dim)
k = torch.randn(batch_size, num_heads, seq_len, head_dim)
v = torch.randn(batch_size, num_heads, seq_len, head_dim)

# Optional mask tensor (causal mask shape)
mask = torch.triu(torch.ones(seq_len, seq_len) * float('-inf'), diagonal=1)

ARGS = {
    "q": q,
    "k": k,
    "v": v,
    "head_dim": head_dim,
    "mask": None,
    "is_causal": True,
    "dropout_p": 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 Tensor, ndim 4, dtype family float, shape (2, 3, 2, 4) — shape (2, 3, 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

Also present in other papers' code

Of 2 samples lifted from this paper's linked repositories, 2 carry a function identity and 0 do not. Of the 2 distinct functions with an identity, 1 are also present in other papers whose code we hold (1 with at least one paper dated earlier, 0 the same month, 0 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
scaled_dot_product_attention 1 other paper
1 dated earlier
2501.03575 (2025-01)
Cosmos World Foundation Model Platform for Physical AI

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

3 Vector Institute Original Lane change (4m) Original Editing Input Dolly zoom Driving simulation Single view Inputs Dynamic video Generated video 0 Cameras 4 8 26 39 106 118 Sparse views 12 Cinematic effect 12 22 42 0 4 8 12 69 69 N Frame number Figure 1. GEN3C can generate long and temporally consistent videos with precise camera control. We apply it to various applications, including single-view and sparse-views novel view synthesis, monocular dynamic video novel view synthesis, and driving simulation. With an explicit 3D cache, GEN3C further supports generating videos with cinematic effects, such as Dolly Zoom which simultaneously changes poses and intrinsics, and 3D editing.

For agents

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

get_harvested_code_for_paper("2503.03751")
get_code_for_paper("2503.03751")
have("2503.03751")

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

curl https://syntology.ai/api/ran/2503.03751.json

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

[![Syntology run record](https://syntology.ai/api/ran/2503.03751.svg)](https://syntology.ai/paper/2503.03751)

Connect an agent — have() is free.