Yunzhi Zhang

Agrim Gupta, Stephen Tian, Yunzhi Zhang et al. · stanford

The ability to predict future visual observations conditioned on past observations and motor commands can enable embodied agents to plan solutions to a variety of tasks in complex environments. This work shows that we can create good video prediction models by pre-training transformers via masked visual modeling. Our approach, named MaskViT, is based on two simple design decisions. First, for memory and training efficiency, we use two types of window attention: spatial and spatiotemporal. Second, during training, we mask a variable percentage of tokens instead of a fixed mask ratio. For inference, MaskViT generates all tokens via iterative refinement where we incrementally decrease the masking ratio following a mask scheduling function. On several datasets we demonstrate that MaskViT outperforms prior works in video prediction, is parameter efficient, and can generate high-resolution videos (256x256). Further, we demonstrate the benefits of inference speedup (up to 512x) due to iterative decoding by using MaskViT for planning on a real robot. Our work suggests that we can endow embodied agents with powerful predictive models by leveraging the general framework of masked visual modeling with minimal domain knowledge.

Kyle Sargent, Zizhang Li, Tanmay Shah et al. · stanford

We introduce a 3D-aware diffusion model, ZeroNVS, for single-image novel view synthesis for in-the-wild scenes. While existing methods are designed for single objects with masked backgrounds, we propose new techniques to address challenges introduced by in-the-wild multi-object scenes with complex backgrounds. Specifically, we train a generative prior on a mixture of data sources that capture object-centric, indoor, and outdoor scenes. To address issues from data mixture such as depth-scale ambiguity, we propose a novel camera conditioning parameterization and normalization scheme. Further, we observe that Score Distillation Sampling (SDS) tends to truncate the distribution of complex backgrounds during distillation of 360-degree scenes, and propose "SDS anchoring" to improve the diversity of synthesized novel views. Our model sets a new state-of-the-art result in LPIPS on the DTU dataset in the zero-shot setting, even outperforming methods specifically trained on DTU. We further adapt the challenging Mip-NeRF 360 dataset as a new benchmark for single-image novel view synthesis, and demonstrate strong performance in this setting. Our code and data are at http://kylesargent.github.io/zeronvs/

Zhengfei Kuang, Yunzhi Zhang, Hong-Xing Yu et al. · stanford

We introduce Stanford-ORB, a new real-world 3D Object inverse Rendering Benchmark. Recent advances in inverse rendering have enabled a wide range of real-world applications in 3D content generation, moving rapidly from research and commercial use cases to consumer devices. While the results continue to improve, there is no real-world benchmark that can quantitatively assess and compare the performance of various inverse rendering methods. Existing real-world datasets typically only consist of the shape and multi-view images of objects, which are not sufficient for evaluating the quality of material recovery and object relighting. Methods capable of recovering material and lighting often resort to synthetic data for quantitative evaluation, which on the other hand does not guarantee generalization to complex real-world environments. We introduce a new dataset of real-world objects captured under a variety of natural scenes with ground-truth 3D scans, multi-view images, and environment lighting. Using this dataset, we establish the first comprehensive real-world evaluation benchmark for object inverse rendering tasks from in-the-wild scenes, and compare the performance of various existing methods.

Ruocheng Wang, Yunzhi Zhang, Jiayuan Mao et al. · stanford

We study the problem of translating an image-based, step-by-step assembly manual created by human designers into machine-interpretable instructions. We formulate this problem as a sequential prediction task: at each step, our model reads the manual, locates the components to be added to the current shape, and infers their 3D poses. This task poses the challenge of establishing a 2D-3D correspondence between the manual image and the real 3D object, and 3D pose estimation for unseen 3D objects, since a new component to be added in a step can be an object built from previous steps. To address these two challenges, we present a novel learning-based framework, the Manual-to-Executable-Plan Network (MEPNet), which reconstructs the assembly steps from a sequence of manual images. The key idea is to integrate neural 2D keypoint detection modules and 2D-3D projection algorithms for high-precision prediction and strong generalization to unseen components. The MEPNet outperforms existing methods on three newly collected LEGO manual datasets and a Minecraft house dataset.

Yunzhi Zhang, Shangzhe Wu, Noah Snavely et al. · deepmind, stanford

What is a rose, visually? A rose comprises its intrinsics, including the distribution of geometry, texture, and material specific to its object category. With knowledge of these intrinsic properties, we may render roses of different sizes and shapes, in different poses, and under different lighting conditions. In this work, we build a generative model that learns to capture such object intrinsics from a single image, such as a photo of a bouquet. Such an image includes multiple instances of an object type. These instances all share the same intrinsics, but appear different due to a combination of variance within these intrinsics and differences in extrinsic factors, such as pose and illumination. Experiments show that our model successfully learns object intrinsics (distribution of geometry, texture, and material) for a wide range of objects, each from a single Internet image. Our method achieves superior results on multiple downstream tasks, including intrinsic image decomposition, shape and image generation, view synthesis, and relighting.

Zizhang Li, Dor Litvak, Ruining Li et al. · oxford, stanford

Learning 3D models of all animals on the Earth requires massively scaling up existing solutions. With this ultimate goal in mind, we develop 3D-Fauna, an approach that learns a pan-category deformable 3D animal model for more than 100 animal species jointly. One crucial bottleneck of modeling animals is the limited availability of training data, which we overcome by simply learning from 2D Internet images. We show that prior category-specific attempts fail to generalize to rare species with limited training images. We address this challenge by introducing the Semantic Bank of Skinned Models (SBSM), which automatically discovers a small set of base animal shapes by combining geometric inductive priors with semantic knowledge implicitly captured by an off-the-shelf self-supervised feature extractor. To train such a model, we also contribute a new large-scale dataset of diverse animal species. At inference time, given a single image of any quadruped animal, our model reconstructs an articulated 3D mesh in a feed-forward fashion within seconds.

Sharon Lee, Yunzhi Zhang, Shangzhe Wu et al. · stanford

Our understanding of the visual world is centered around various concept axes, characterizing different aspects of visual entities. While different concept axes can be easily specified by language, e.g. color, the exact visual nuances along each axis often exceed the limitations of linguistic articulations, e.g. a particular style of painting. In this work, our goal is to learn a language-informed visual concept representation, by simply distilling large pre-trained vision-language models. Specifically, we train a set of concept encoders to encode the information pertinent to a set of language-informed concept axes, with an objective of reproducing the input image through a pre-trained Text-to-Image (T2I) model. To encourage better disentanglement of different concept encoders, we anchor the concept embeddings to a set of text embeddings obtained from a pre-trained Visual Question Answering (VQA) model. At inference time, the model extracts concept embeddings along various axes from new test images, which can be remixed to generate images with novel compositions of visual concepts. With a lightweight test-time finetuning procedure, it can also generalize to novel concepts unseen at training.

Keqiang Sun, Dor Litvak, Yunzhi Zhang et al.

We introduce a new method for learning a generative model of articulated 3D animal motions from raw, unlabeled online videos. Unlike existing approaches for 3D motion synthesis, our model requires no pose annotations or parametric shape models for training; it learns purely from a collection of unlabeled web video clips, leveraging semantic correspondences distilled from self-supervised image features. At the core of our method is a video Photo-Geometric Auto-Encoding framework that decomposes each training video clip into a set of explicit geometric and photometric representations, including a rest-pose 3D shape, an articulated pose sequence, and texture, with the objective of re-rendering the input video via a differentiable renderer. This decomposition allows us to learn a generative model over the underlying articulated pose sequences akin to a Variational Auto-Encoding (VAE) formulation, but without requiring any external pose annotations. At inference time, we can generate new motion sequences by sampling from the learned motion VAE, and create plausible 4D animations of an animal automatically within seconds given a single input image.

Yunzhi Zhang, Zizhang Li, Amit Raj et al. · stanford

We propose 3D Congealing, a novel problem of 3D-aware alignment for 2D images capturing semantically similar objects. Given a collection of unlabeled Internet images, our goal is to associate the shared semantic parts from the inputs and aggregate the knowledge from 2D images to a shared 3D canonical space. We introduce a general framework that tackles the task without assuming shape templates, poses, or any camera parameters. At its core is a canonical 3D representation that encapsulates geometric and semantic information. The framework optimizes for the canonical representation together with the pose for each input image, and a per-image coordinate map that warps 2D pixel coordinates to the 3D canonical frame to account for the shape matching. The optimization procedure fuses prior knowledge from a pre-trained image generative model and semantic information from input images. The former provides strong knowledge guidance for this under-constraint task, while the latter provides the necessary information to mitigate the training data bias from the pre-trained model. Our framework can be used for various tasks such as correspondence matching, pose estimation, and image editing, achieving strong results on real-world image datasets under challenging illumination conditions and on in-the-wild online image collections.

Andreas Engelhardt, Amit Raj, Mark Boss et al.

We present SHINOBI, an end-to-end framework for the reconstruction of shape, material, and illumination from object images captured with varying lighting, pose, and background. Inverse rendering of an object based on unconstrained image collections is a long-standing challenge in computer vision and graphics and requires a joint optimization over shape, radiance, and pose. We show that an implicit shape representation based on a multi-resolution hash encoding enables faster and robust shape reconstruction with joint camera alignment optimization that outperforms prior work. Further, to enable the editing of illumination and object reflectance (i.e. material) we jointly optimize BRDF and illumination together with the object's shape. Our method is class-agnostic and works on in-the-wild image collections of objects to produce relightable 3D assets for several use cases such as AR/VR, movies, games, etc. Project page: https://shinobi.aengelhardt.com Video: https://www.youtube.com/watch?v=iFENQ6AcYd8&feature=youtu.be

Yunzhi Zhang, Jiajun Wu

Novel view synthesis (NVS) and video prediction (VP) are typically considered disjoint tasks in computer vision. However, they can both be seen as ways to observe the spatial-temporal world: NVS aims to synthesize a scene from a new point of view, while VP aims to see a scene from a new point of time. These two tasks provide complementary signals to obtain a scene representation, as viewpoint changes from spatial observations inform depth, and temporal observations inform the motion of cameras and individual objects. Inspired by these observations, we propose to study the problem of Video Extrapolation in Space and Time (VEST). We propose a model that leverages the self-supervision and the complementary cues from both tasks, while existing methods can only solve one of them. Experiments show that our method achieves performance better than or comparable to several state-of-the-art NVS and VP methods on indoor and outdoor real-world datasets.

Wilson Yan, Yunzhi Zhang, Pieter Abbeel et al.

We present VideoGPT: a conceptually simple architecture for scaling likelihood based generative modeling to natural videos. VideoGPT uses VQ-VAE that learns downsampled discrete latent representations of a raw video by employing 3D convolutions and axial self-attention. A simple GPT-like architecture is then used to autoregressively model the discrete latents using spatio-temporal position encodings. Despite the simplicity in formulation and ease of training, our architecture is able to generate samples competitive with state-of-the-art GAN models for video generation on the BAIR Robot dataset, and generate high fidelity natural videos from UCF-101 and Tumbler GIF Dataset (TGIF). We hope our proposed architecture serves as a reproducible reference for a minimalistic implementation of transformer based video generation models. Samples and code are available at https://wilson1yan.github.io/videogpt/index.html

Yunzhi Zhang, Pieter Abbeel, Lerrel Pinto

Continually solving new, unsolved tasks is the key to learning diverse behaviors. Through reinforcement learning (RL), we have made massive strides towards solving tasks that have a single goal. However, in the multi-task domain, where an agent needs to reach multiple goals, the choice of training goals can largely affect sample efficiency. When biological agents learn, there is often an organized and meaningful order to which learning happens. Inspired by this, we propose setting up an automatic curriculum for goals that the agent needs to solve. Our key insight is that if we can sample goals at the frontier of the set of goals that an agent is able to reach, it will provide a significantly stronger learning signal compared to randomly sampled goals. To operationalize this idea, we introduce a goal proposal module that prioritizes goals that maximize the epistemic uncertainty of the Q-function of the policy. This simple technique samples goals that are neither too hard nor too easy for the agent to solve, hence enabling continual improvement. We evaluate our method across 13 multi-goal robotic tasks and 5 navigation tasks, and demonstrate performance gains over current state-of-the-art methods.

Yunzhi Zhang, Ignasi Clavera, Boren Tsai et al.

Significant progress has been made in the area of model-based reinforcement learning. State-of-the-art algorithms are now able to match the asymptotic performance of model-free methods while being significantly more data efficient. However, this success has come at a price: state-of-the-art model-based methods require significant computation interleaved with data collection, resulting in run times that take days, even if the amount of agent interaction might be just hours or even minutes. When considering the goal of learning in real-time on real robots, this means these state-of-the-art model-based algorithms still remain impractical. In this work, we propose an asynchronous framework for model-based reinforcement learning methods that brings down the run time of these algorithms to be just the data collection time. We evaluate our asynchronous framework on a range of standard MuJoCo benchmarks. We also evaluate our asynchronous framework on three real-world robotic manipulation tasks. We show how asynchronous learning not only speeds up learning w.r.t wall-clock time through parallelization, but also further reduces the sample complexity of model-based approaches by means of improving the exploration and by means of effectively avoiding the policy overfitting to the deficiencies of learned dynamics models.