Pith. sign in

REVIEW 11 cited by

FitVid: Overfitting in Pixel-Level Video Prediction

Not yet reviewed by Pith; the record is open.

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2106.13195 v1 pith:ZREX7JIJ submitted 2021-06-24 cs.CV cs.LG

FitVid: Overfitting in Pixel-Level Video Prediction

classification cs.CV cs.LG
keywords videomodelsbenchmarkscurrentfitvidoverfittingpredictionquality
verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved
0 comments
read the original abstract

An agent that is capable of predicting what happens next can perform a variety of tasks through planning with no additional training. Furthermore, such an agent can internally represent the complex dynamics of the real-world and therefore can acquire a representation useful for a variety of visual perception tasks. This makes predicting the future frames of a video, conditioned on the observed past and potentially future actions, an interesting task which remains exceptionally challenging despite many recent advances. Existing video prediction models have shown promising results on simple narrow benchmarks but they generate low quality predictions on real-life datasets with more complicated dynamics or broader domain. There is a growing body of evidence that underfitting on the training data is one of the primary causes for the low quality predictions. In this paper, we argue that the inefficient use of parameters in the current video models is the main reason for underfitting. Therefore, we introduce a new architecture, named FitVid, which is capable of severe overfitting on the common benchmarks while having similar parameter count as the current state-of-the-art models. We analyze the consequences of overfitting, illustrating how it can produce unexpected outcomes such as generating high quality output by repeating the training data, and how it can be mitigated using existing image augmentation techniques. As a result, FitVid outperforms the current state-of-the-art models across four different video prediction benchmarks on four different metrics.

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Forward citations

Cited by 11 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score.

  1. Zero-Shot Robotic Manipulation with Pretrained Image-Editing Diffusion Models

    cs.RO 2023-10 conditional novelty 7.0

    SuSIE uses a finetuned InstructPix2Pix diffusion model to propose subgoal images that guide a low-level goal-conditioned policy, achieving SOTA zero-shot performance on CALVIN and real-world manipulation.

  2. Language Model Beats Diffusion -- Tokenizer is Key to Visual Generation

    cs.CV 2023-10 unverdicted novelty 7.0

    A new shared video-image tokenizer enables large language models to surpass diffusion models on standard visual generation benchmarks.

  3. Phenaki: Variable Length Video Generation From Open Domain Textual Description

    cs.CV 2022-10 unverdicted novelty 7.0

    Phenaki generates arbitrary-length videos from sequences of text prompts by tokenizing videos with causal temporal attention and generating tokens with a text-conditioned masked transformer, trained jointly on images ...

  4. Imagen Video: High Definition Video Generation with Diffusion Models

    cs.CV 2022-10 unverdicted novelty 7.0

    Imagen Video generates high-definition text-conditional videos via a cascade of base and super-resolution diffusion models, achieving high fidelity and controllability.

  5. Video Diffusion Models

    cs.CV 2022-04 unverdicted novelty 7.0

    A diffusion model for video generation extends image architectures with joint image-video training and improved conditional sampling, delivering first large-scale text-to-video results and state-of-the-art performance...

  6. Stochastic Lifting for Generating Trajectories of Stochastic Physical Systems

    cs.LG 2026-05 unverdicted novelty 6.0

    Stochastic Lifting adds random labels to training transitions to train a regression model that generates diverse stochastic trajectories without collapsing to mean predictions.

  7. Inferring Dynamic Physical Properties from Video Foundation Models

    cs.CV 2025-10 unverdicted novelty 6.0

    Video foundation models infer dynamic physical properties such as elasticity, viscosity, and friction from videos at levels close to classical oracles while outperforming current MLLMs with suitable prompting.

  8. Long-Context Autoregressive Video Modeling with Next-Frame Prediction

    cs.CV 2025-03 unverdicted novelty 6.0

    FAR baseline plus asymmetric kernels for long short-term context modeling achieves SOTA short and long video generation in autoregressive setups.

  9. MagicVideo: Efficient Video Generation With Latent Diffusion Models

    cs.CV 2022-11 unverdicted novelty 6.0

    MagicVideo generates 256x256 text-conditioned video clips via latent diffusion with a custom 3D U-Net, achieving roughly 64 times lower compute than prior video diffusion models.

  10. Geometry-Aware Single-Image 4D Synthesis via Dense Trajectory Generation

    cs.CV 2025-12 conditional novelty 5.0

    A diffusion model generates dense 4D point trajectories from a single image, and a separate view-synthesis module renders them into novel-view videos.

  11. Bridge-WA: Predicting Where and How the World Changes for Robotic Action

    cs.RO 2026-07 unverdicted novelty 4.0

    Bridge-WA introduces a lightweight distillation-based world-action model that uses future-change priors to improve robotic task success and robustness without deployment-time dense rollouts.