X-MIC: Cross-Modal Instance Conditioning for Egocentric Action Generalization

Lately, there has been growing interest in adapting vision-language models (VLMs) to image and third-person video classification due to their success in zero-shot recognition. However, the adaptation of these models to egocentric videos has been largely unexplored. To address this gap, we propose a simple yet effective cross-modal adaptation framework, which we call X-MIC. Using a video adapter, our pipeline learns to align frozen text embeddings to each egocentric video directly in the shared embedding space. Our novel adapter architecture retains and improves generalization of the pre-trained VLMs by disentangling learnable temporal modeling and frozen visual encoder. This results in an enhanced alignment of text embeddings to each egocentric video, leading to a significant improvement in cross-dataset generalization. We evaluate our approach on the Epic-Kitchens, Ego4D, and EGTEA datasets for fine-grained cross-dataset action generalization, demonstrating the effectiveness of our method. Code is available at https://github.com/annusha/xmic

DiffH2O: Diffusion-Based Synthesis of Hand-Object Interactions from Textual Descriptions

Generating natural hand-object interactions in 3D is challenging as the resulting hand and object motions are expected to be physically plausible and semantically meaningful. Furthermore, generalization to unseen objects is hindered by the limited scale of available hand-object interaction datasets. We propose DiffH2O, a novel method to synthesize realistic, one or two-handed object interactions from provided text prompts and geometry of the object. The method introduces three techniques that enable effective learning from limited data. First, we decompose the task into a grasping stage and a text-based interaction stage and use separate diffusion models for each. In the grasping stage, the model only generates hand motions, whereas in the interaction phase both hand and object poses are synthesized. Second, we propose a compact representation that tightly couples hand and object poses. Third, we propose two different guidance schemes to allow more control of the generated motions: grasp guidance and detailed textual guidance. Grasp guidance takes a single target grasping pose and guides the diffusion model to reach this grasp at the end of the grasping stage, which provides control over the grasping pose. Given a grasping motion from this stage, multiple different actions can be prompted in the interaction phase. For textual guidance, we contribute comprehensive text descriptions to the GRAB dataset and show that they enable our method to have more fine-grained control over hand-object interactions. Our quantitative and qualitative evaluation demonstrates that the proposed method outperforms baseline methods and leads to natural hand-object motions. Moreover, we demonstrate the practicality of our framework by utilizing a hand pose estimate from an off-the-shelf pose estimator for guidance, and then sampling multiple different actions in the interaction stage.

FoundPose: Unseen Object Pose Estimation with Foundation Features

We propose FoundPose, a method for 6D pose estimation of unseen rigid objects from a single RGB image. The method assumes that 3D models of the objects are available but does not require any object-specific training. This is achieved by building upon DINOv2, a recent vision foundation model with impressive generalization capabilities. An online pose estimation stage is supported by a minimal object representation that is built during a short onboarding stage from DINOv2 patch features extracted from rendered object templates. Given a query image with an object segmentation mask, FoundPose first rapidly retrieves a handful of similarly looking templates by a DINOv2-based bag-of-words approach. Pose hypotheses are then generated from 2D-3D correspondences established by matching DINOv2 patch features between the query image and a retrieved template, and finally optimized by featuremetric refinement. The method can handle diverse objects, including challenging ones with symmetries and without any texture, and noticeably outperforms existing RGB methods for coarse pose estimation in both accuracy and speed on the standard BOP benchmark. With the featuremetric and additional MegaPose refinement, which are demonstrated complementary, the method outperforms all RGB competitors. Source code is at: evinpinar.github.io/foundpose.

HoloAssist: an Egocentric Human Interaction Dataset for Interactive AI Assistants in the Real World

Building an interactive AI assistant that can perceive, reason, and collaborate with humans in the real world has been a long-standing pursuit in the AI community. This work is part of a broader research effort to develop intelligent agents that can interactively guide humans through performing tasks in the physical world. As a first step in this direction, we introduce HoloAssist, a large-scale egocentric human interaction dataset, where two people collaboratively complete physical manipulation tasks. The task performer executes the task while wearing a mixed-reality headset that captures seven synchronized data streams. The task instructor watches the performer's egocentric video in real time and guides them verbally. By augmenting the data with action and conversational annotations and observing the rich behaviors of various participants, we present key insights into how human assistants correct mistakes, intervene in the task completion procedure, and ground their instructions to the environment. HoloAssist spans 166 hours of data captured by 350 unique instructor-performer pairs. Furthermore, we construct and present benchmarks on mistake detection, intervention type prediction, and hand forecasting, along with detailed analysis. We expect HoloAssist will provide an important resource for building AI assistants that can fluidly collaborate with humans in the real world. Data can be downloaded at https://holoassist.github.io/.

Context-Aware Sequence Alignment using 4D Skeletal Augmentation

Temporal alignment of fine-grained human actions in videos is important for numerous applications in computer vision, robotics, and mixed reality. State-of-the-art methods directly learn image-based embedding space by leveraging powerful deep convolutional neural networks. While being straightforward, their results are far from satisfactory, the aligned videos exhibit severe temporal discontinuity without additional post-processing steps. The recent advancements in human body and hand pose estimation in the wild promise new ways of addressing the task of human action alignment in videos. In this work, based on off-the-shelf human pose estimators, we propose a novel context-aware self-supervised learning architecture to align sequences of actions. We name it CASA. Specifically, CASA employs self-attention and cross-attention mechanisms to incorporate the spatial and temporal context of human actions, which can solve the temporal discontinuity problem. Moreover, we introduce a self-supervised learning scheme that is empowered by novel 4D augmentation techniques for 3D skeleton representations. We systematically evaluate the key components of our method. Our experiments on three public datasets demonstrate CASA significantly improves phase progress and Kendall's Tau scores over the previous state-of-the-art methods.

Learning to Align Sequential Actions in the Wild

State-of-the-art methods for self-supervised sequential action alignment rely on deep networks that find correspondences across videos in time. They either learn frame-to-frame mapping across sequences, which does not leverage temporal information, or assume monotonic alignment between each video pair, which ignores variations in the order of actions. As such, these methods are not able to deal with common real-world scenarios that involve background frames or videos that contain non-monotonic sequence of actions. In this paper, we propose an approach to align sequential actions in the wild that involve diverse temporal variations. To this end, we propose an approach to enforce temporal priors on the optimal transport matrix, which leverages temporal consistency, while allowing for variations in the order of actions. Our model accounts for both monotonic and non-monotonic sequences and handles background frames that should not be aligned. We demonstrate that our approach consistently outperforms the state-of-the-art in self-supervised sequential action representation learning on four different benchmark datasets.

Reconstructing and grounding narrated instructional videos in 3D

Narrated instructional videos often show and describe manipulations of similar objects, e.g., repairing a particular model of a car or laptop. In this work we aim to reconstruct such objects and to localize associated narrations in 3D. Contrary to the standard scenario of instance-level 3D reconstruction, where identical objects or scenes are present in all views, objects in different instructional videos may have large appearance variations given varying conditions and versions of the same product. Narrations may also have large variation in natural language expressions. We address these challenges by three contributions. First, we propose an approach for correspondence estimation combining learnt local features and dense flow. Second, we design a two-step divide and conquer reconstruction approach where the initial 3D reconstructions of individual videos are combined into a 3D alignment graph. Finally, we propose an unsupervised approach to ground natural language in obtained 3D reconstructions. We demonstrate the effectiveness of our approach for the domain of car maintenance. Given raw instructional videos and no manual supervision, our method successfully reconstructs engines of different car models and associates textual descriptions with corresponding objects in 3D.

H2O: Two Hands Manipulating Objects for First Person Interaction Recognition

We present, for the first time, a comprehensive framework for egocentric interaction recognition using markerless 3D annotations of two hands manipulating objects. To this end, we propose a method to create a unified dataset for egocentric 3D interaction recognition. Our method produces annotations of the 3D pose of two hands and the 6D pose of the manipulated objects, along with their interaction labels for each frame. Our dataset, called H2O (2 Hands and Objects), provides synchronized multi-view RGB-D images, interaction labels, object classes, ground-truth 3D poses for left & right hands, 6D object poses, ground-truth camera poses, object meshes and scene point clouds. To the best of our knowledge, this is the first benchmark that enables the study of first-person actions with the use of the pose of both left and right hands manipulating objects and presents an unprecedented level of detail for egocentric 3D interaction recognition. We further propose the first method to predict interaction classes by estimating the 3D pose of two hands and the 6D pose of the manipulated objects, jointly from RGB images. Our method models both inter- and intra-dependencies between both hands and objects by learning the topology of a graph convolutional network that predicts interactions. We show that our method facilitated by this dataset establishes a strong baseline for joint hand-object pose estimation and achieves state-of-the-art accuracy for first person interaction recognition.

HoloLens 2 Research Mode as a Tool for Computer Vision Research

Mixed reality headsets, such as the Microsoft HoloLens 2, are powerful sensing devices with integrated compute capabilities, which makes it an ideal platform for computer vision research. In this technical report, we present HoloLens 2 Research Mode, an API and a set of tools enabling access to the raw sensor streams. We provide an overview of the API and explain how it can be used to build mixed reality applications based on processing sensor data. We also show how to combine the Research Mode sensor data with the built-in eye and hand tracking capabilities provided by HoloLens 2. By releasing the Research Mode API and a set of open-source tools, we aim to foster further research in the fields of computer vision as well as robotics and encourage contributions from the research community.

Leveraging Photometric Consistency over Time for Sparsely Supervised Hand-Object Reconstruction

Modeling hand-object manipulations is essential for understanding how humans interact with their environment. While of practical importance, estimating the pose of hands and objects during interactions is challenging due to the large mutual occlusions that occur during manipulation. Recent efforts have been directed towards fully-supervised methods that require large amounts of labeled training samples. Collecting 3D ground-truth data for hand-object interactions, however, is costly, tedious, and error-prone. To overcome this challenge we present a method to leverage photometric consistency across time when annotations are only available for a sparse subset of frames in a video. Our model is trained end-to-end on color images to jointly reconstruct hands and objects in 3D by inferring their poses. Given our estimated reconstructions, we differentiably render the optical flow between pairs of adjacent images and use it within the network to warp one frame to another. We then apply a self-supervised photometric loss that relies on the visual consistency between nearby images. We achieve state-of-the-art results on 3D hand-object reconstruction benchmarks and demonstrate that our approach allows us to improve the pose estimation accuracy by leveraging information from neighboring frames in low-data regimes.