Human behavior forecasting during human-human interactions is of utmost importance to provide robotic or virtual agents with social intelligence. This problem is especially challenging for scenarios that are highly driven by interpersonal dynamics. In this work, we present the first systematic comparison of state-of-the-art approaches for behavior forecasting. To do so, we leverage whole-body annotations (face, body, and hands) from the very recently released UDIVA v0.5, which features face-to-face dyadic interactions. Our best attention-based approaches achieve state-of-the-art performance in UDIVA v0.5. We show that by autoregressively predicting the future with methods trained for the short-term future (<400ms), we outperform the baselines even for a considerably longer-term future (up to 2s). We also show that this finding holds when highly noisy annotations are used, which opens new horizons towards the use of weakly-supervised learning. Combined with large-scale datasets, this may help boost the advances in this field.
Non-verbal social human behavior forecasting has increasingly attracted the interest of the research community in recent years. Its direct applications to human-robot interaction and socially-aware human motion generation make it a very attractive field. In this survey, we define the behavior forecasting problem for multiple interactive agents in a generic way that aims at unifying the fields of social signals prediction and human motion forecasting, traditionally separated. We hold that both problem formulations refer to the same conceptual problem, and identify many shared fundamental challenges: future stochasticity, context awareness, history exploitation, etc. We also propose a taxonomy that comprises methods published in the last 5 years in a very informative way and describes the current main concerns of the community with regard to this problem. In order to promote further research on this field, we also provide a summarised and friendly overview of audiovisual datasets featuring non-acted social interactions. Finally, we describe the most common metrics used in this task and their particular issues.
This paper reports the results and post-challenge analyses of ChaLearn's AutoDL challenge series, which helped sorting out a profusion of AutoML solutions for Deep Learning (DL) that had been introduced in a variety of settings, but lacked fair comparisons. All input data modalities (time series, images, videos, text, tabular) were formatted as tensors and all tasks were multi-label classification problems. Code submissions were executed on hidden tasks, with limited time and computational resources, pushing solutions that get results quickly. In this setting, DL methods dominated, though popular Neural Architecture Search (NAS) was impractical. Solutions relied on fine-tuned pre-trained networks, with architectures matching data modality. Post-challenge tests did not reveal improvements beyond the imposed time limit. While no component is particularly original or novel, a high level modular organization emerged featuring a "meta-learner", "data ingestor", "model selector", "model/learner", and "evaluator". This modularity enabled ablation studies, which revealed the importance of (off-platform) meta-learning, ensembling, and efficient data management. Experiments on heterogeneous module combinations further confirm the (local) optimality of the winning solutions. Our challenge legacy includes an ever-lasting benchmark (http://autodl.chalearn.org), the open-sourced code of the winners, and a free "AutoDL self-service".
It has been a long history that most object detection methods obtain objects by using the non-maximum suppression (NMS) and its improved versions like Soft-NMS to remove redundant bounding boxes. We challenge those NMS-based methods from three aspects: 1) The bounding box with highest confidence value may not be the true positive having the biggest overlap with the ground-truth box. 2) Not only suppression is required for redundant boxes, but also confidence enhancement is needed for those true positives. 3) Sorting candidate boxes by confidence values is not necessary so that full parallelism is achievable. In this paper, inspired by belief propagation (BP), we propose the Confidence Propagation Cluster (CP-Cluster) to replace NMS-based methods, which is fully parallelizable as well as better in accuracy. In CP-Cluster, we borrow the message passing mechanism from BP to penalize redundant boxes and enhance true positives simultaneously in an iterative way until convergence. We verified the effectiveness of CP-Cluster by applying it to various mainstream detectors such as FasterRCNN, SSD, FCOS, YOLOv3, YOLOv5, Centernet etc. Experiments on MS COCO show that our plug and play method, without retraining detectors, is able to steadily improve average mAP of all those state-of-the-art models with a clear margin from 0.2 to 1.9 respectively when compared with NMS-based methods. Source code is available at https://github.com/shenyi0220/CP-Cluster
This paper aims to reduce the rendering time of generalizable radiance fields. Some recent works equip neural radiance fields with image encoders and are able to generalize across scenes, which avoids the per-scene optimization. However, their rendering process is generally very slow. A major factor is that they sample lots of points in empty space when inferring radiance fields. In this paper, we present a hybrid scene representation which combines the best of implicit radiance fields and explicit depth maps for efficient rendering. Specifically, we first build the cascade cost volume to efficiently predict the coarse geometry of the scene. The coarse geometry allows us to sample few points near the scene surface and significantly improves the rendering speed. This process is fully differentiable, enabling us to jointly learn the depth prediction and radiance field networks from only RGB images. Experiments show that the proposed approach exhibits state-of-the-art performance on the DTU, Real Forward-facing and NeRF Synthetic datasets, while being at least 50 times faster than previous generalizable radiance field methods. We also demonstrate the capability of our method to synthesize free-viewpoint videos of dynamic human performers in real-time. The code will be available at https://zju3dv.github.io/enerf/.
Obtaining standardized crowdsourced benchmark of computational methods is a major issue in scientific communities. Dedicated frameworks enabling fair continuous benchmarking in a unified environment are yet to be developed. Here we introduce Codabench, an open-sourced, community-driven platform for benchmarking algorithms or software agents versus datasets or tasks. A public instance of Codabench is open to everyone, free of charge, and allows benchmark organizers to compare fairly submissions, under the same setting (software, hardware, data, algorithms), with custom protocols and data formats. Codabench has unique features facilitating the organization of benchmarks flexibly, easily and reproducibly. Firstly, it supports code submission and data submission for testing on dedicated compute workers, which can be supplied by the benchmark organizers. This makes the system scalable, at low cost for the platform providers. Secondly, Codabench benchmarks are created from self-contained bundles, which are zip files containing a full description of the benchmark in a configuration file (following a well-defined schema), documentation pages, data, ingestion and scoring programs, making benchmarks reusable and portable. The Codabench documentation includes many examples of bundles that can serve as templates. Thirdly, Codabench uses dockers for each task's running environment to make results reproducible. Codabench has been used internally and externally with more than 10 applications during the past 6 months. As illustrative use cases, we introduce 4 diverse benchmarks covering Graph Machine Learning, Cancer Heterogeneity, Clinical Diagnosis and Reinforcement Learning.
Analyzing better time series with limited human effort is of interest to academia and industry. Driven by business scenarios, we organized the first Automated Time Series Regression challenge (AutoSeries) for the WSDM Cup 2020. We present its design, analysis, and post-hoc experiments. The code submission requirement precluded participants from any manual intervention, testing automated machine learning capabilities of solutions, across many datasets, under hardware and time limitations. We prepared 10 datasets from diverse application domains (sales, power consumption, air quality, traffic, and parking), featuring missing data, mixed continuous and categorical variables, and various sampling rates. Each dataset was split into a training and a test sequence (which was streamed, allowing models to continuously adapt). The setting of time series regression, differs from classical forecasting in that covariates at the present time are known. Great strides were made by participants to tackle this AutoSeries problem, as demonstrated by the jump in performance from the sample submission, and post-hoc comparisons with AutoGluon. Simple yet effective methods were used, based on feature engineering, LightGBM, and random search hyper-parameter tuning, addressing all aspects of the challenge. Our post-hoc analyses revealed that providing additional time did not yield significant improvements. The winners' code was open-sourced https://www.4paradigm.com/competition/autoseries2020.
Machine learning has recently demonstrated impressive progress in predictive accuracy across a wide array of tasks. Most ML approaches focus on generalization performance on unseen data that are similar to the training data (In-Distribution, or IND). However, real world applications and deployments of ML rarely enjoy the comfort of encountering examples that are always IND. In such situations, most ML models commonly display erratic behavior on Out-of-Distribution (OOD) examples, such as assigning high confidence to wrong predictions, or vice-versa. Implications of such unusual model behavior are further exacerbated in the healthcare setting, where patient health can potentially be put at risk. It is crucial to study the behavior and robustness properties of models under distributional shift, understand common failure modes, and take mitigation steps before the model is deployed. Having a benchmark that shines light upon these aspects of a model is a first and necessary step in addressing the issue. Recent work and interest in increasing model robustness in OOD settings have focused more on image modality, while the Electronic Health Record (EHR) modality is still largely under-explored. We aim to bridge this gap by releasing BEDS-Bench, a benchmark for quantifying the behavior of ML models over EHR data under OOD settings. We use two open access, de-identified EHR datasets to construct several OOD data settings to run tests on, and measure relevant metrics that characterize crucial aspects of a model's OOD behavior. We evaluate several learning algorithms under BEDS-Bench and find that all of them show poor generalization performance under distributional shift in general. Our results highlight the need and the potential to improve robustness of EHR models under distributional shift, and BEDS-Bench provides one way to measure progress towards that goal.