# Inserting Videos into Videos

arXiv: https://arxiv.org/abs/1903.06571 · HTML: https://mlsanity.com/paper/1903.06571/

## Key facts

- Cited by 7 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 15 Mar 2019.

## Built on by (most cited first, 7 of 7)

- [GeoSim: Realistic Video Simulation via Geometry-Aware Composition for Self-Driving](https://mlsanity.com/paper/2101.06543/) (2021)
- [Spatial-Separated Curve Rendering Network for Efficient and High-Resolution Image Harmonization](https://mlsanity.com/paper/2109.05750/) (2021)
- [LEMaRT: Label-Efficient Masked Region Transform for Image Harmonization](https://mlsanity.com/paper/2304.13166/) (2023)
- [Anything in Any Scene: Photorealistic Video Object Insertion](https://mlsanity.com/paper/2401.17509/) (2024)
- [DreamInsert: Zero-Shot Image-to-Video Object Insertion from A Single Image](https://mlsanity.com/paper/2503.10342/) (2025)
- [Animating Street View](https://mlsanity.com/paper/2310.08534/) (2023)
- [Image Harmonization with Region-wise Contrastive Learning](https://mlsanity.com/paper/2205.14058/) (2022)

## Abstract (arXiv)

In this paper, we introduce a new problem of manipulating a given video by inserting other videos into it. Our main task is, given an object video and a scene video, to insert the object video at a user-specified location in the scene video so that the resulting video looks realistic. We aim to handle different object motions and complex backgrounds without expensive segmentation annotations. As it is difficult to collect training pairs for this problem, we synthesize fake training pairs that can provide helpful supervisory signals when training a neural network with unpaired real data. The proposed network architecture can take both real and fake pairs as input and perform both supervised and unsupervised training in an adversarial learning scheme. To synthesize a realistic video, the network renders each frame based on the current input and previous frames. Within this framework, we observe that injecting noise into previous frames while generating the current frame stabilizes training. We conduct experiments on real-world videos in object tracking and person re-identification benchmark datasets. Experimental results demonstrate that the proposed algorithm is able to synthesize long sequences of realistic videos with a given object video inserted.