We propose a method for editing images from human instructions: given an input image and a written instruction that tells the model what to do, our model follows these instructions to edit the image. To obtain training data for this problem, we combine the knowledge of two large pretrained models—a language model (GPT-3) and a text-to-image model (Stable Diffusion)—to generate a large dataset of image editing examples. Our conditional diffusion model, InstructPix2Pix, is trained on our generated data, and generalizes to real images and user-written instructions at inference time. Since it performs edits in the forward pass and does not require per-example fine-tuning or inversion, our model edits images quickly, in a matter of seconds. We show compelling editing results for a diverse collection of input images and written instructions.

We study the problem of inferring scene affordances by presenting a method for realistically inserting people into scenes. Given a scene image with a marked region and an image of a person, we insert the person into the scene while respecting the scene affordances. Our model can infer the set of realistic poses given the scene context, re-pose the reference person, and harmonize the composition. We set up the task in a self-supervised fashion by learning to repose humans in video clips. We train a large-scale diffusion model on a dataset of 2.4M video clips that produces diverse plausible poses while respecting the scene context. Given the learned human-scene composition, our model can also hallucinate realistic people and scenes when prompted without conditioning and also enables interactive editing. A quantitative evaluation shows that our method synthesizes more realistic human appearance and more natural human-scene interactions than prior work.

We present a video generation model that accurately reproduces object motion, changes in camera viewpoint, and new content that arises over time. Existing video generation methods often fail to produce new content as a function of time while maintaining consistencies expected in real environments, such as plausible dynamics and object persistence. A common failure case is for content to never change due to over-reliance on inductive bias to provide temporal consistency, such as a single latent code that dictates content for the entire video. On the other extreme, without long-term consistency, generated videos may morph unrealistically between different scenes. To address these limitations, we prioritize the time axis by redesigning the temporal latent representation and learning long-term consistency from data by training on longer videos. We leverage a two-phase training strategy, where we separately train using longer videos at a low resolution and shorter videos at a high resolution. To evaluate the capabilities of our model, we introduce two new benchmark datasets with explicit focus on long-term temporal dynamics.

We present a technique for synthesizing a motion blurred image from a pair of unblurred images captured in succession. To build this system we motivate and design a differentiable “line prediction” layer to be used as part of a neural network architecture, with which we can learn a system to regress from image pairs to motion blurred images that span the capture time of the input image pair. Training this model requires an abundance of data, and so we design and execute a strategy for using frame interpolation techniques to generate a large-scale synthetic dataset of motion blurred images and their respective inputs. We additionally capture a high quality test set of real motion blurred images, synthesized from slow motion videos, with which we evaluate our model against several baseline techniques that can be used to synthesize motion blur. Our model produces higher accuracy output than our baselines, and is several orders of magnitude faster than baselines with competitive accuracy.

Machine learning techniques work best when the data used for training resembles the data used for evaluation. This holds true for learned single-image denoising algorithms, which are applied to real raw camera sensor readings but, due to practical constraints, are often trained on synthetic image data. Though it is understood that generalizing from synthetic to real images requires careful consideration of the noise properties of camera sensors, the other aspects of an image processing pipeline (such as gain, color correction, and tone mapping) are often overlooked, despite their significant effect on how raw measurements are transformed into finished images. To address this, we present a technique to “unprocess” images by inverting each step of an image processing pipeline, thereby allowing us to synthesize realistic raw sensor measurements from commonly available Internet photos. We additionally model the relevant components of an image processing pipeline when evaluating our loss function, which allows training to be aware of all relevant photometric processing that will occur after denoising. By unprocessing and processing training data and model outputs in this way, we are able to train a simple convolutional neural network that has 14%-38% lower error rates and is 9×-18× faster than the previous state of the art on the Darmstadt Noise Dataset, and generalizes to sensors outside of that dataset as well.

Taking photographs in low light using a mobile phone is challenging and rarely produces pleasing results. Aside from the physical limits imposed by read noise and photon shot noise, these cameras are typically handheld, have small apertures and sensors, use mass-produced analog electronics that cannot easily be cooled, and are commonly used to photograph subjects that move, like children and pets. In this paper we describe a system for capturing clean, sharp, colorful photographs in light as low as 0.3 lux, where human vision becomes monochromatic and indistinct. To permit handheld photography without flash illumination, we capture, align, and combine multiple frames. Our system employs "motion metering", which uses an estimate of motion magnitudes (whether due to handshake or moving objects) to identify the number of frames and the per-frame exposure times that together minimize both noise and motion blur in a captured burst. We combine these frames using robust alignment and merging techniques that are specialized for high-noise imagery. To ensure accurate colors in such low light, we employ a learning-based auto white balancing algorithm. To prevent the photographs from looking like they were shot in daylight, we use tone mapping techniques inspired by illusionistic painting: increasing contrast, crushing shadows to black, and surrounding the scene with darkness. All of these processes are performed using the limited computational resources of a mobile device. Our system can be used by novice photographers to produce shareable pictures in a few seconds based on a single shutter press, even in environments so dim that humans cannot see clearly.

The SoundsRight Audio Completely Automated Public Turing tests to tell Computers and Humans Apart (CAPTCHA) was developed with the goal of providing a usable and secure audio CAPTCHA for people with visual impairments. Its design requires users to repeatedly identify a specific sound from a group of different sounds (e.g. baby crying and bird chirping) in real time. Adding background noise (sound masks) to the sounds may make it more difficult for automated software to recognise the sounds and therefore, improve security. However, the sound masks may also make it more challenging for human users to recognise the sound. We conducted a user study involving 20 blind participants and 20 sighted participants to investigate the effect of sound masks on the usability of the SoundsRight CAPTCHA. The results suggest that sound masks do have a significant impact on the failure rate and response time. Sighted participants had significantly a higher failure rate than blind participants and were more vulnerable to the negative effect of sound masks.

In this paper we describe the development of a new audio CAPTCHA called the SoundsRight CAPTCHA, and the evaluation of the CAPTCHA with 20 blind users. Blind users cannot use visual CAPTCHAs, and it has been documented in the research literature that the existing audio CAPTCHAs have task success rates below 50% for blind users. The SoundsRight audio CAPTCHA presents a real-time audio-based challenge in which the user is asked to identify a specific sound (for example the sound of a bell or a piano) each time it occurs in a series of 10 sounds that are played through the computer's audio system. Evaluation results from three rounds of usability testing document that the task success rate was higher than 90% for blind users. Discussion, limitations, and suggestions for future research are also presented.