Motion control is crucial for generating expressive and compelling video content; however, most existing video generation models rely mainly on text prompts for control, which struggle to capture the nuances of dynamic actions and temporal compositions. To this end, we train a video generation model conditioned on spatiotemporally sparse or dense motion trajectories. In contrast to prior motion conditioning work, this flexible representation can encode any number of trajectories, object-specific or global scene motion, and temporally sparse motion; due to its flexibility we refer to this conditioning as motion prompts. While users may directly specify sparse trajectories, we also show how to translate high-level user requests into detailed, semi-dense motion prompts, a process we term motion prompt expansion. We demonstrate the versatility of our approach through various applications, including camera and object motion control, "interacting" with an image, motion transfer, and image editing. Our results showcase emergent behaviors, such as realistic physics, suggesting the potential of motion prompts for probing video models and interacting with future generative world models. Finally, we evaluate quantitatively, conduct a human study, and demonstrate strong performance.

We address the problem of synthesizing multi-view optical illusions: images that change appearance upon a transformation, such as a flip or rotation. We propose a simple, zero-shot method for obtaining these illusions from off-the-shelf text-to-image diffusion models. During the reverse diffusion process, we estimate the noise from different views of a noisy image, and then combine these noise estimates together and denoise the image. A theoretical analysis suggests that this method works precisely for views that can be written as orthogonal transformations, of which permutations are a subset. This leads to the idea of a visual anagram-an image that changes appearance under some rearrangement of pixels. This includes rotations and flips, but also more exotic pixel permutations such as a jigsaw rearrangement. Our approach also naturally extends to illusions with more than two views. We provide both qualitative and quantitative results demonstrating the effectiveness and flexibility of our method. Please see our project webpage for additional visualizations and results: https://dangeng.github.io/visual_anagrams/.

Diffusion models are capable of generating impressive images conditioned on text descriptions, and extensions of these models allow users to edit images at a relatively coarse scale. However, the ability to precisely edit the layout, position, pose, and shape of objects in images with diffusion models is still difficult. To this end, we propose _motion guidance_, a zero-shot technique that allows a user to specify dense, complex motion fields that indicate where each pixel in an image should move. Motion guidance works by steering the diffusion sampling process with the gradients through an off-the-shelf optical flow network. Specifically, we design a guidance loss that encourages the sample to have the desired motion, as estimated by a flow network, while also being visually similar to the source image. By simultaneously sampling from a diffusion model and guiding the sample to have low guidance loss, we can obtain a motion-edited image. We demonstrate that our technique works on complex motions and produces high quality edits of real and generated images.

Spectrograms are 2D representations of sound that look very different from the images found in our visual world. And natural images, when played as spectrograms, make unnatural sounds. In this paper, we show that it is possible to synthesize spectrograms that simultaneously look like natural images and sound like natural audio. We call these visual spectrograms *images that sound*. Our approach is simple and zero-shot, and it leverages pre-trained text-to-image and text-to-spectrogram diffusion models that operate in a shared latent space. During the reverse process, we denoise noisy latents with both the audio and image diffusion models in parallel, resulting in a sample that is likely under both models. Through quantitative evaluations and perceptual studies, we find that our method successfully generates spectrograms that align with a desired audio prompt while also taking the visual appearance of a desired image prompt.

This paper presents a simple, self-supervised method for magnifying subtle motions in video: given an input video and a magnification factor, we manipulate the video such that its new optical flow is scaled by the desired amount. To train our model, we propose a loss function that estimates the optical flow of the generated video and penalizes how far if deviates from the given magnification factor. Thus, training involves differentiating through a pretrained optical flow network. Since our model is self-supervised, we can further improve its performance through test-time adaptation, by finetuning it on the input video. It can also be easily extended to magnify the motions of only user-selected objects. Our approach avoids the need for synthetic magnification datasets that have been used to train prior learning-based approaches. Instead, it leverages the existing capabilities of off-the-shelf motion estimators. We demonstrate the effectiveness of our method through evaluations of both visual quality and quantitative metrics on a range of real-world and synthetic videos, and we show our method works for both supervised and unsupervised optical flow methods.

Image prediction methods often struggle on tasks that require changing the positions of objects, such as video prediction, producing blurry images that average over the many positions that objects might occupy. In this paper, we propose a simple change to existing image similarity metrics that makes them more robust to positional errors: we match the images using optical flow, then measure the visual similarity of corresponding pixels. This change leads to crisper and more perceptually accurate predictions, and does not require modifications to the image prediction network. We apply our method to a variety of video prediction tasks, where it obtains strong performance with simple network architectures, and to the closely related task of video interpolation. Code and results are available at our webpage: https://dangeng.github.io/CorrWiseLosses

Autonomous agents situated in real-world environments must be able to master large repertoires of skills. While a single short skill can be learned quickly, it would be impractical to learn every task independently. Instead, the agent should share knowledge across behaviors such that each task can be learned efficiently, and such that the resulting model can generalize to new tasks, especially ones that are compositions or subsets of tasks seen previously. A policy conditioned on a goal or demonstration has the potential to share knowledge between tasks if it sees enough diversity of inputs. However, these methods may not generalize to a more complex task at test time. We introduce compositional plan vectors (CPVs) to enable a policy to perform compositions of tasks without additional supervision. CPVs represent trajectories as the sum of the subtasks within them. We show that CPVs can be learned within a one-shot imitation learning framework without any additional supervision or information about task hierarchy, and enable a demonstration-conditioned policy to generalize to tasks that sequence twice as many skills as the tasks seen during training. Analogously to embeddings such as word2vec in NLP, CPVs can also support simple arithmetic operations -- for example, we can add the CPVs for two different tasks to command an agent to compose both tasks, without any additional training.