Closing the Loop: Training-Free Revisit Consistency for Autoregressive Generative Rendering

Hugging Face Daily Papers Papers
generative-rendering autoregressive-models video-generation 3d-rendering consistency training-free computer-vision

Summary

This paper introduces a training-free method to improve revisit consistency in autoregressive generative rendering by using temporal and spatial correspondences from the 3D engine to maintain consistent appearance when the camera revisits locations.

Recent conditional video generation models have shown promising potentials to transform 3D engine renderings, such as depth maps and untextured geometry, into photorealistic videos for gaming and immersive content creation. These applications require long-horizon auto-regressive generation that continuously synthesizes new frames while preserving a persistent 3D world. Auto-regressive generators synthesize video chunk by chunk with a bounded KV cache, so when the camera revisits a location after its context has been evicted, the model often regenerates inconsistent appearance, even though the conditioning renderings (e.g., depth) remain perfectly aligned with the underlying geometry.We address this revisit inconsistency without any post-training by exploiting correspondences the 3D engine already provides: temporal correspondence retrieves pose-matched historical latent chunks into the KV cache as loop-closure memory, while spatial correspondence from camera pose and depth reprojection biases token-level attention toward geometrically corresponding regions of the retrieved chunks. We demonstrate our method on loop-closure trajectories mined from TartanAir and TartanGround dataset to mirror complicate real-world application scenarios, where it outperforms existing training-free baselines on revisit consistency without losing overall video quality. Project Page: https://wenchao-m.github.io/ClosetheLoop.github.io/
Original Article
View Cached Full Text

Cached at: 07/27/26, 05:40 AM

Paper page - Closing the Loop: Training-Free Revisit Consistency for Autoregressive Generative Rendering

Source: https://huggingface.co/papers/2607.21848

Abstract

Recentconditionalvideogenerationmodelshaveshownpromisingpotentialstotransform3Denginerenderings,suchasdepthmapsanduntexturedgeometry,intophotorealisticvideosforgamingandimmersivecontentcreation.Theseapplicationsrequirelong-horizonauto-regressivegenerationthatcontinuouslysynthesizesnewframeswhilepreservingapersistent3Dworld.Auto-regressivegeneratorssynthesizevideochunkbychunkwithaboundedKVcache,sowhenthecamerarevisitsalocationafteritscontexthasbeenevicted,themodeloftenregeneratesinconsistentappearance,eventhoughtheconditioningrenderings(e.g.,depth)remainperfectlyalignedwiththeunderlyinggeometry.Weaddressthisrevisitinconsistencywithoutanypost-trainingbyexploitingcorrespondencesthe3Denginealreadyprovides:temporalcorrespondenceretrievespose-matchedhistoricallatentchunksintotheKVcacheasloop-closurememory,whilespatialcorrespondencefromcameraposeanddepthreprojectionbiasestoken-levelattentiontowardgeometricallycorrespondingregionsoftheretrievedchunks.Wedemonstrateourmethodonloop-closuretrajectoriesminedfromTartanAirandTartanGrounddatasettomirrorcomplicatereal-worldapplicationscenarios,whereitoutperformsexistingtraining-freebaselinesonrevisitconsistencywithoutlosingoverallvideoquality.ProjectPage:https://wenchao-m.github.io/ClosetheLoop.github.io/

View arXiv pageView PDFProject pageAdd to collection

Get this paper in your agent:

hf papers read 2607\.21848

Don’t have the latest CLI?curl \-LsSf https://hf\.co/cli/install\.sh \| bash

Models citing this paper0

No model linking this paper

Cite arxiv.org/abs/2607.21848 in a model README.md to link it from this page.

Datasets citing this paper0

No dataset linking this paper

Cite arxiv.org/abs/2607.21848 in a dataset README.md to link it from this page.

Spaces citing this paper0

No Space linking this paper

Cite arxiv.org/abs/2607.21848 in a Space README.md to link it from this page.

Collections including this paper0

No Collection including this paper

Add this paper to acollectionto link it from this page.

Similar Articles

Consistency Models

OpenAI Blog

OpenAI introduces Consistency Models, a new family of generative models that enable fast one-step image generation by directly mapping noise to data, while supporting multi-step sampling and zero-shot editing tasks like inpainting and super-resolution. The approach achieves state-of-the-art FID scores on CIFAR-10 and ImageNet 64x64 for one-step generation.