While language tasks are naturally expressed in a single, unified, modeling framework, i.e., generating sequences of tokens, this has not been the case in computer vision. As a result, there is a proliferation of distinct architectures and loss functions for different vision tasks. In this work we show that a diverse set of "core" computer vision tasks can also be unified if formulated in terms of a shared pixel-to-sequence interface. We focus on four tasks, namely, object detection, instance segmentation, keypoint detection, and image captioning, all with diverse types of outputs, e.g., bounding boxes or dense masks. Despite that, by formulating the output of each task as a sequence of discrete tokens with a unified interface, we show that one can train a neural network with a single model architecture and loss function on all these tasks, with no task-specific customization. To solve a specific task, we use a short prompt as task description, and the sequence output adapts to the prompt so it can produce task-specific output. We show that such a model can achieve competitive performance compared to well-established task-specific models.
Coordinate networks like Multiplicative Filter Networks (MFNs) and BACON offer some control over the frequency spectrum used to represent continuous signals such as images or 3D volumes. Yet, they are not readily applicable to problems for which coarse-to-fine estimation is required, including various inverse problems in which coarse-to-fine optimization plays a key role in avoiding poor local minima. We introduce a new coordinate network architecture and training scheme that enables coarse-to-fine optimization with fine-grained control over the frequency support of learned reconstructions. This is achieved with two key innovations. First, we incorporate skip connections so that structure at one scale is preserved when fitting finer-scale structure. Second, we propose a novel initialization scheme to provide control over the model frequency spectrum at each stage of optimization. We demonstrate how these modifications enable multiscale optimization for coarse-to-fine fitting to natural images. We then evaluate our model on synthetically generated datasets for the the problem of single-particle cryo-EM reconstruction. We learn high resolution multiscale structures, on par with the state-of-the art.
Generating temporally coherent high fidelity video is an important milestone in generative modeling research. We make progress towards this milestone by proposing a diffusion model for video generation that shows very promising initial results. Our model is a natural extension of the standard image diffusion architecture, and it enables jointly training from image and video data, which we find to reduce the variance of minibatch gradients and speed up optimization. To generate long and higher resolution videos we introduce a new conditional sampling technique for spatial and temporal video extension that performs better than previously proposed methods. We present the first results on a large text-conditioned video generation task, as well as state-of-the-art results on an established unconditional video generation benchmark. Supplementary material is available at https://video-diffusion.github.io/
Data is the driving force of machine learning, with the amount and quality of training data often being more important for the performance of a system than architecture and training details. But collecting, processing and annotating real data at scale is difficult, expensive, and frequently raises additional privacy, fairness and legal concerns. Synthetic data is a powerful tool with the potential to address these shortcomings: 1) it is cheap 2) supports rich ground-truth annotations 3) offers full control over data and 4) can circumvent or mitigate problems regarding bias, privacy and licensing. Unfortunately, software tools for effective data generation are less mature than those for architecture design and training, which leads to fragmented generation efforts. To address these problems we introduce Kubric, an open-source Python framework that interfaces with PyBullet and Blender to generate photo-realistic scenes, with rich annotations, and seamlessly scales to large jobs distributed over thousands of machines, and generating TBs of data. We demonstrate the effectiveness of Kubric by presenting a series of 13 different generated datasets for tasks ranging from studying 3D NeRF models to optical flow estimation. We release Kubric, the used assets, all of the generation code, as well as the rendered datasets for reuse and modification.
We introduce Palette, a simple and general framework for image-to-image translation using conditional diffusion models. On four challenging image-to-image translation tasks (colorization, inpainting, uncropping, and JPEG decompression), Palette outperforms strong GAN and regression baselines, and establishes a new state of the art. This is accomplished without task-specific hyper-parameter tuning, architecture customization, or any auxiliary loss, demonstrating a desirable degree of generality and flexibility. We uncover the impact of using $L_2$ vs. $L_1$ loss in the denoising diffusion objective on sample diversity, and demonstrate the importance of self-attention through empirical architecture studies. Importantly, we advocate a unified evaluation protocol based on ImageNet, and report several sample quality scores including FID, Inception Score, Classification Accuracy of a pre-trained ResNet-50, and Perceptual Distance against reference images for various baselines. We expect this standardized evaluation protocol to play a critical role in advancing image-to-image translation research. Finally, we show that a single generalist Palette model trained on 3 tasks (colorization, inpainting, JPEG decompression) performs as well or better than task-specific specialist counterparts.
This paper presents Pix2Seq, a simple and generic framework for object detection. Unlike existing approaches that explicitly integrate prior knowledge about the task, we simply cast object detection as a language modeling task conditioned on the observed pixel inputs. Object descriptions (e.g., bounding boxes and class labels) are expressed as sequences of discrete tokens, and we train a neural net to perceive the image and generate the desired sequence. Our approach is based mainly on the intuition that if a neural net knows about where and what the objects are, we just need to teach it how to read them out. Beyond the use of task-specific data augmentations, our approach makes minimal assumptions about the task, yet it achieves competitive results on the challenging COCO dataset, compared to highly specialized and well optimized detection algorithms.
We present SR3, an approach to image Super-Resolution via Repeated Refinement. SR3 adapts denoising diffusion probabilistic models to conditional image generation and performs super-resolution through a stochastic denoising process. Inference starts with pure Gaussian noise and iteratively refines the noisy output using a U-Net model trained on denoising at various noise levels. SR3 exhibits strong performance on super-resolution tasks at different magnification factors, on faces and natural images. We conduct human evaluation on a standard 8X face super-resolution task on CelebA-HQ, comparing with SOTA GAN methods. SR3 achieves a fool rate close to 50%, suggesting photo-realistic outputs, while GANs do not exceed a fool rate of 34%. We further show the effectiveness of SR3 in cascaded image generation, where generative models are chained with super-resolution models, yielding a competitive FID score of 11.3 on ImageNet.
Adversarial robustness is an open challenge in deep learning, most often tackled using adversarial training. Adversarial training is computationally costly, involving alternated optimization with a trade-off between standard generalization and adversarial robustness. We explore training robust models without adversarial training by revisiting a known result linking maximally robust classifiers and minimum norm solutions, and combining it with recent results on the implicit bias of optimizers. First, we show that, under certain conditions, it is possible to achieve both perfect standard accuracy and a certain degree of robustness without a trade-off, simply by training an overparameterized model using the implicit bias of the optimization. In that regime, there is a direct relationship between the type of the optimizer and the attack to which the model is robust. Second, we investigate the role of the architecture in designing robust models. In particular, we characterize the robustness of linear convolutional models, showing that they resist attacks subject to a constraint on the Fourier-$\ell_\infty$ norm. This result explains the property of $\ell_p$-bounded adversarial perturbations that tend to be concentrated in the Fourier domain. This leads us to a novel attack in the Fourier domain that is inspired by the well-known frequency-dependent sensitivity of human perception. We evaluate Fourier-$\ell_\infty$ robustness of recent CIFAR-10 models with robust training and visualize adversarial perturbations.
Capsule networks are designed to parse an image into a hierarchy of objects, parts and relations. While promising, they remain limited by an inability to learn effective low level part descriptions. To address this issue we propose a novel self-supervised method for learning part descriptors of an image. During training, we exploit motion as a powerful perceptual cue for part definition, using an expressive decoder for part generation and layered image formation with occlusion. Experiments demonstrate robust part discovery in the presence of multiple objects, cluttered backgrounds, and significant occlusion. The resulting part descriptors, a.k.a. part capsules, are decoded into shape masks, filling in occluded pixels, along with relative depth on single images. We also report unsupervised object classification using our capsule parts in a stacked capsule autoencoder.
The impact of gradient noise on training deep models is widely acknowledged but not well understood. In this context, we study the distribution of gradients during training. We introduce a method, Gradient Clustering, to minimize the variance of average mini-batch gradient with stratified sampling. We prove that the variance of average mini-batch gradient is minimized if the elements are sampled from a weighted clustering in the gradient space. We measure the gradient variance on common deep learning benchmarks and observe that, contrary to common assumptions, gradient variance increases during training, and smaller learning rates coincide with higher variance. In addition, we introduce normalized gradient variance as a statistic that better correlates with the speed of convergence compared to gradient variance.