Tactile DreamFusion: Exploiting Tactile Sensing for 3D Generation

3D generation methods have shown visually compelling results powered by diffusion image priors. However, they often fail to produce realistic geometric details, resulting in overly smooth surfaces or geometric details inaccurately baked in albedo maps. To address this, we introduce a new method that incorporates touch as an additional modality to improve the geometric details of generated 3D assets. We design a lightweight 3D texture field to synthesize visual and tactile textures, guided by 2D diffusion model priors on both visual and tactile domains. We condition the visual texture generation on high-resolution tactile normals and guide the patch-based tactile texture refinement with a customized TextureDreambooth. We further present a multi-part generation pipeline that enables us to synthesize different textures across various regions. To our knowledge, we are the first to leverage high-resolution tactile sensing to enhance geometric details for 3D generation tasks. We evaluate our method in both text-to-3D and image-to-3D settings. Our experiments demonstrate that our method provides customized and realistic fine geometric textures while maintaining accurate alignment between two modalities of vision and touch.

FlashTex: Fast Relightable Mesh Texturing with LightControlNet

Manually creating textures for 3D meshes is time-consuming, even for expert visual content creators. We propose a fast approach for automatically texturing an input 3D mesh based on a user-provided text prompt. Importantly, our approach disentangles lighting from surface material/reflectance in the resulting texture so that the mesh can be properly relit and rendered in any lighting environment. We introduce LightControlNet, a new text-to-image model based on the ControlNet architecture, which allows the specification of the desired lighting as a conditioning image to the model. Our text-to-texture pipeline then constructs the texture in two stages. The first stage produces a sparse set of visually consistent reference views of the mesh using LightControlNet. The second stage applies a texture optimization based on Score Distillation Sampling (SDS) that works with LightControlNet to increase the texture quality while disentangling surface material from lighting. Our pipeline is significantly faster than previous text-to-texture methods, while producing high-quality and relightable textures.

StableLego: Stability Analysis of Block Stacking Assembly

Recent advancements in robotics enable robots to accomplish complex assembly tasks. However, designing an assembly requires a non-trivial effort since a slight variation in the design could significantly affect the task feasibility. It is critical to ensure the physical feasibility of the assembly design so that the assembly task can be successfully executed. To address the challenge, this paper studies the physical stability of assembly structures, in particular, block stacking assembly, where people use cubic blocks to build 3D structures (e.g., Lego constructions). The paper proposes a new optimization formulation, which optimizes over force balancing equations, for inferring the structural stability of 3D block-stacking structures. The proposed stability analysis is tested and verified on hand-crafted Lego examples. The experiment results demonstrate that the proposed stability analysis can correctly predict whether the structure is stable. In addition, it outperforms the existing methods since it can locate the weakest parts in the design, and more importantly, solve any given assembly structure. To further validate the proposed analysis formulation, we provide StableLego: a comprehensive dataset including more than 50k 3D objects with their Lego layouts. We test the proposed stability analysis and include the stability inference for each corresponding object in StableLego. Our code and the dataset are available at https://github.com/intelligent-control-lab/StableLego.

Total-Recon: Deformable Scene Reconstruction for Embodied View Synthesis

We explore the task of embodied view synthesis from monocular videos of deformable scenes. Given a minute-long RGBD video of people interacting with their pets, we render the scene from novel camera trajectories derived from in-scene motion of actors: (1) egocentric cameras that simulate the point of view of a target actor and (2) 3rd-person cameras that follow the actor. Building such a system requires reconstructing the root-body and articulated motion of each actor in the scene, as well as a scene representation that supports free-viewpoint synthesis. Longer videos are more likely to capture the scene from diverse viewpoints (which helps reconstruction) but are also more likely to contain larger motions (which complicates reconstruction). To address these challenges, we present Total-Recon, the first method to photorealistically reconstruct deformable scenes from long monocular RGBD videos. Crucially, to scale to long videos, our method hierarchically decomposes the scene motion into the motion of each object, which itself is decomposed into global root-body motion and local articulations. To quantify such "in-the-wild" reconstruction and view synthesis, we collect ground-truth data from a specialized stereo RGBD capture rig for 11 challenging videos, significantly outperforming prior art. Code, videos, and data can be found at https://andrewsonga.github.io/totalrecon .

3D-aware Conditional Image Synthesis

We propose pix2pix3D, a 3D-aware conditional generative model for controllable photorealistic image synthesis. Given a 2D label map, such as a segmentation or edge map, our model learns to synthesize a corresponding image from different viewpoints. To enable explicit 3D user control, we extend conditional generative models with neural radiance fields. Given widely-available monocular images and label map pairs, our model learns to assign a label to every 3D point in addition to color and density, which enables it to render the image and pixel-aligned label map simultaneously. Finally, we build an interactive system that allows users to edit the label map from any viewpoint and generate outputs accordingly.

Depth-supervised NeRF: Fewer Views and Faster Training for Free

One common failure mode of Neural Radiance Field (NeRF) models is fitting incorrect geometries when given an insufficient number of input views. We propose DS-NeRF (Depth-supervised Neural Radiance Fields), a loss for learning neural radiance fields that takes advantage of readily-available depth supervision. Our key insight is that sparse depth supervision can be used to regularize the learned geometry, a crucial component for effectively rendering novel views using NeRF. We exploit the fact that current NeRF pipelines require images with known camera poses that are typically estimated by running structure-from-motion (SFM). Crucially, SFM also produces sparse 3D points that can be used as ``free" depth supervision during training: we simply add a loss to ensure that depth rendered along rays that intersect these 3D points is close to the observed depth. We find that DS-NeRF can render more accurate images given fewer training views while training 2-6x faster. With only two training views on real-world images, DS-NeRF significantly outperforms NeRF as well as other sparse-view variants. We show that our loss is compatible with these NeRF models, demonstrating that depth is a cheap and easily digestible supervisory signal. Finally, we show that DS-NeRF supports other types of depth supervision such as scanned depth sensors and RGBD reconstruction outputs.

Unsupervised Any-to-Many Audiovisual Synthesis via Exemplar Autoencoders

We present an unsupervised approach that enables us to convert the speech input of any one individual to an output set of potentially-infinitely many speakers. One can stand in front of a mic and be able to make their favorite celebrity say the same words. Our approach builds on simple autoencoders that project out-of-sample data to the distribution of the training set (motivated by PCA/linear autoencoders). We use an exemplar autoencoder to learn the voice and specific style (emotions and ambiance) of a target speaker. In contrast to existing methods, the proposed approach can be easily extended to an arbitrarily large number of speakers in a very little time using only two-three minutes of audio data from a speaker. We also exhibit the usefulness of our approach for generating video from audio signals and vice-versa. We suggest the reader to check out our project webpage for various synthesized examples: https://dunbar12138.github.io/projectpage/Audiovisual/