Geometric models like DUSt3R have shown great advances in understanding the geometry of a scene from pairs of photos. However, they fail when the inputs are from vastly different viewpoints (e.g., aerial vs.\ ground) or modalities (e.g., photos vs.\ abstract drawings) compared to what was observed during training. This paper addresses a challenging version of this problem: predicting correspondences between ground-level photos and floor plans. Current datasets for joint photo--floor plan reasoning are limited, either lacking in varying modalities (VIGOR) or lacking in correspondences (WAFFLE). To address these limitations, we introduce a new dataset, C3, created by first reconstructing a number of scenes in 3D from Internet photo collections via structure-from-motion, then manually registering the reconstructions to floor plans gathered from the Internet, from which we can derive correspondence between images and floor plans. C3 contains 90K paired floor plans and photos across 597 scenes with 153M pixel-level correspondences and 85K camera poses. We find that state-of-the-art correspondence models struggle on this task. By training on our new data, we can improve on the best performing method by 34\% in RMSE. We also identify open challenges in cross-modal geometric reasoning that our dataset aims to help address. Our project website is available at: \url{https://c3po-correspondence.github.io/}.

In this paper, we are interested in modeling a how-to instructional procedure, such as a cooking recipe, with a meaningful and rich high-level representation. Specifically, we propose to represent cooking recipes and food images as cooking programs. Programs provide a structured repre-sentation of the task, capturing cooking semantics and se-quential relationships of actions in the form of a graph. This allows them to be easily manipulated by users and executed by agents. To this end, we build a model that is trained to learn a joint embedding between recipes and food images via self-supervision and jointly generate a program from this embedding as a sequence. To validate our idea, we crowdsource programs for cooking recipes and show that: (a) projecting the image-recipe embeddings into programs leads to better cross-modal retrieval results; (b) generating programs from images leads to better recognition re-sults compared to predicting raw cooking instructions; and (c) we can generate food images by manipulating programs via optimizing the latent code of a GAN. Code, data, and models are available online11http://cookingprograms.csail.mit.edu.

We propose a scalable cooperative control approach which coordinates a group of rigidly connected autonomous surface vessels to track desired trajectories in a planar water environment as a single floating modular structure. Our approach leverages the implicit information of the structure's motion for force and torque allocation without explicit communication among the robots. In our system, a leader robot steers the entire group by adjusting its force and torque according to the structure's deviation from the desired trajectory, while follower robots run distributed consensus-based controllers to match their inputs to amplify the leader's intent using only onboard sensors as feedback. To cope with the nonlinear system dynamics in the water, the leader robot employs a nonlinear model predictive controller (NMPC), where we experimentally estimated the dynamics model of the floating modular structure in order to achieve superior performance for leader-following control. Our method has a wide range of potential applications in transporting humans and goods in many of today's existing waterways. We conducted trajectory and orientation tracking experiments in hardware with three custom-built autonomous modular robotic boats, called Roboat, which are capable of holonomic motions and onboard state estimation. Simulation results with up to 65 robots also prove the scalability of our proposed approach.