Interactions between people are often governed by their relationships. On the flip side, social relationships are built upon several interactions. Two strangers are more likely to greet and introduce themselves while becoming friends over time. We are fascinated by this interplay between interactions and relationships, and believe that it is an important aspect of understanding social situations. In this work, we propose neural models to learn and jointly predict interactions, relationships, and the pair of characters that are involved. We note that interactions are informed by a mixture of visual and dialog cues, and present a multimodal architecture to extract meaningful information from them. Localizing the pair of interacting characters in video is a time-consuming process, instead, we train our model to learn from clip-level weak labels. We evaluate our models on the MovieGraphs dataset and show the impact of modalities, use of longer temporal context for predicting relationships, and achieve encouraging performance using weak labels as compared with ground-truth labels. Code is online.
Annotating videos is cumbersome, expensive and not scalable. Yet, many strong video models still rely on manually annotated data. With the recent introduction of the HowTo100M dataset, narrated videos now offer the possibility of learning video representations without manual supervision. In this work we propose a new learning approach, MIL-NCE, capable of addressing misalignments inherent to narrated videos. With this approach we are able to learn strong video representations from scratch, without the need for any manual annotation. We evaluate our representations on a wide range of four downstream tasks over eight datasets: action recognition (HMDB-51, UCF-101, Kinetics-700), text-to-video retrieval (YouCook2, MSR-VTT), action localization (YouTube-8M Segments, CrossTask) and action segmentation (COIN). Our method outperforms all published self-supervised approaches for these tasks as well as several fully supervised baselines.
We present a method to learn a representation for adverbs from instructional videos using weak supervision from the accompanying narrations. Key to our method is the fact that the visual representation of the adverb is highly dependant on the action to which it applies, although the same adverb will modify multiple actions in a similar way. For instance, while 'spread quickly' and 'mix quickly' will look dissimilar, we can learn a common representation that allows us to recognize both, among other actions. We formulate this as an embedding problem, and use scaled dot-product attention to learn from weakly-supervised video narrations. We jointly learn adverbs as invertible transformations operating on the embedding space, so as to add or remove the effect of the adverb. As there is no prior work on weakly supervised learning from adverbs, we gather paired action-adverb annotations from a subset of the HowTo100M dataset for 6 adverbs: quickly/slowly, finely/coarsely, and partially/completely. Our method outperforms all baselines for video-to-adverb retrieval with a performance of 0.719 mAP. We also demonstrate our model's ability to attend to the relevant video parts in order to determine the adverb for a given action.
Our goal in this work is to improve the performance of human action recognition for viewpoints unseen during training by using synthetic training data. Although synthetic data has been shown to be beneficial for tasks such as human pose estimation, its use for RGB human action recognition is relatively unexplored. We make use of the recent advances in monocular 3D human body reconstruction from real action sequences to automatically render synthetic training videos for the action labels. We make the following contributions: (i) we investigate the extent of variations and augmentations that are beneficial to improving performance at new viewpoints. We consider changes in body shape and clothing for individuals, as well as more action relevant augmentations such as non-uniform frame sampling, and interpolating between the motion of individuals performing the same action; (ii) We introduce a new dataset, SURREACT, that allows supervised training of spatio-temporal CNNs for action classification; (iii) We substantially improve the state-of-the-art action recognition performance on the NTU RGB+D and UESTC standard human action multi-view benchmarks; Finally, (iv) we extend the augmentation approach to in-the-wild videos from a subset of the Kinetics dataset to investigate the case when only one-shot training data is available, and demonstrate improvements in this case as well.
Manipulation tasks such as preparing a meal or assembling furniture remain highly challenging for robotics and vision. The supervised approach of imitation learning can handle short tasks but suffers from compounding errors and the need of many demonstrations for longer and more complex tasks. Reinforcement learning (RL) can find solutions beyond demonstrations but requires tedious and task-specific reward engineering for multi-step problems. In this work we address the difficulties of both methods and explore their combination. To this end, we propose a RL policies operating on pre-trained skills, that can learn composite manipulations using no intermediate rewards and no demonstrations of full tasks. We also propose an efficient training of basic skills from few synthetic demonstrated trajectories by exploring recent CNN architectures and data augmentation. We show successful learning of policies for composite manipulation tasks such as making a simple breakfast. Notably, our method achieves high success rates on a real robot, while using synthetic training data only.
Learning text-video embeddings usually requires a dataset of video clips with manually provided captions. However, such datasets are expensive and time consuming to create and therefore difficult to obtain on a large scale. In this work, we propose instead to learn such embeddings from video data with readily available natural language annotations in the form of automatically transcribed narrations. The contributions of this work are three-fold. First, we introduce HowTo100M: a large-scale dataset of 136 million video clips sourced from 1.22M narrated instructional web videos depicting humans performing and describing over 23k different visual tasks. Our data collection procedure is fast, scalable and does not require any additional manual annotation. Second, we demonstrate that a text-video embedding trained on this data leads to state-of-the-art results for text-to-video retrieval and action localization on instructional video datasets such as YouCook2 or CrossTask. Finally, we show that this embedding transfers well to other domains: fine-tuning on generic Youtube videos (MSR-VTT dataset) and movies (LSMDC dataset) outperforms models trained on these datasets alone. Our dataset, code and models will be publicly available at: www.di.ens.fr/willow/research/howto100m/.
In this paper, we address the problem of visually guided rearrangement planning with many movable objects, i.e., finding a sequence of actions to move a set of objects from an initial arrangement to a desired one, while relying directly on visual inputs coming from a camera. We introduce an efficient and scalable rearrangement planning method, addressing a fundamental limitation of most existing approaches that do not scale well with the number of objects. This increased efficiency allows us to use planning in a closed loop with visual workspace analysis to build a robust rearrangement framework that can recover from errors and external perturbations. The contributions of this work are threefold. First, we develop an AlphaGo-like strategy for rearrangement planning, improving the efficiency of Monte-Carlo Tree Search (MCTS) using a policy trained from rearrangement planning examples. We show empirically that the proposed approach scales well with the number of objects. Second, in order to demonstrate the efficiency of the planner on a real robot, we adopt a state-of-the-art calibration-free visual recognition system that outputs position of a single object and extend it to estimate the state of a workspace containing multiple objects. Third, we validate the complete pipeline with several experiments on a real UR-5 robotic arm solving rearrangement planning problems with multiple movable objects and only requiring few seconds of computation to compute the plan. We also show empirically that the robot can successfully recover from errors and perturbations in the workspace. Source code and pretrained models for our work are available at https://github.com/ylabbe/rearrangement-planning
Metric Learning for visual similarity has mostly adopted binary supervision indicating whether a pair of images are of the same class or not. Such a binary indicator covers only a limited subset of image relations, and is not sufficient to represent semantic similarity between images described by continuous and/or structured labels such as object poses, image captions, and scene graphs. Motivated by this, we present a novel method for deep metric learning using continuous labels. First, we propose a new triplet loss that allows distance ratios in the label space to be preserved in the learned metric space. The proposed loss thus enables our model to learn the degree of similarity rather than just the order. Furthermore, we design a triplet mining strategy adapted to metric learning with continuous labels. We address three different image retrieval tasks with continuous labels in terms of human poses, room layouts and image captions, and demonstrate the superior performance of our approach compared to previous methods.