We evaluate existing foundation models video understanding capabilities using a carefully designed experiment protocol consisting of three hallmark tasks (action recognition, temporal localization, and spatiotemporal localization), eight datasets well received by the community, and four adaptation methods tailoring a foundation model (FM) for a downstream task. Moreover, we propose a scalar VideoGLUE score (VGS) to measure an FMs efficacy and efficiency when adapting to general video understanding tasks. Our main findings are as follows. First, task-specialized models significantly outperform the six FMs studied in this work, in sharp contrast to what FMs have achieved in natural language and image understanding. Second,video-native FMs, whose pretraining data contains the video modality, are generally better than image-native FMs in classifying motion-rich videos, localizing actions in time, and understanding a video of more than one action. Third, the video-native FMs can perform well on video tasks under light adaptations to downstream tasks(e.g., freezing the FM backbones), while image-native FMs win in full end-to-end finetuning. The first two observations reveal the need and tremendous opportunities to conduct research on video-focused FMs, and the last confirms that both tasks and adaptation methods matter when it comes to the evaluation of FMs. Our code is released under: https://github.com/tensorflow/models/tree/master/official/projects/videoglue.

We introduce Imagen 3, a latent diffusion model that generates high quality images from text prompts. We describe our quality and responsibility evaluations. Imagen 3 is preferred over other state-of-the-art (SOTA) models at the time of evaluation. In addition, we discuss issues around safety and representation, as well as methods we used to minimize the potential harm of our models.

Computational approaches leveraging computer vision and machine learning have transformed the quantification of animal behavior from video. However, existing methods often rely on task-specific features or models, which struggle to generalize across diverse datasets and tasks. Recent advances in machine learning, particularly the emergence of vision foundation models, i.e., large-scale models pre-trained on massive, diverse visual repositories, offers a way to tackle these challenges. Here, we investigate the potential of frozen video foundation models across a range of behavior analysis tasks, including classification, retrieval, and localization. We use a single, frozen model to extract general-purpose representations from video data, and perform extensive evaluations on diverse open-sourced animal behavior datasets. Our results demonstrate that features with minimal adaptation from foundation models achieve competitive performance compared to existing methods specifically designed for each dataset, across species, behaviors, and experimental contexts. This highlights the potential of frozen video foundation models as a powerful and accessible backbone for automated behavior analysis, with the ability to accelerate research across diverse fields from neuroscience, to ethology, and to ecology. ### Competing Interest Statement The authors have declared no competing interest.

Existing video-language pre-training methods primarily focus on instance-level alignment between video clips and captions via global contrastive learning but neglect rich fine-grained local information in both videos and text, which is of importance to downstream tasks requiring temporal localization and semantic reasoning. A powerful model is expected to be capable of capturing region-object correspondences and recognizing scene changes in a video clip, reflecting spatial and temporal granularity, respectively. To strengthen model's understanding into such fine-grained details, we propose a simple yet effective video-language modeling framework, S-ViLM, by exploiting the intrinsic structures of these two modalities. It includes two novel designs, inter-clip spatial grounding and intra-clip temporal grouping, to promote learning region-object alignment and temporal-aware features, simultaneously. Comprehensive evaluations demonstrate that S-ViLM performs favorably against existing approaches in learning more expressive representations. Specifically, S-ViLM surpasses the state-of-the-art methods substantially on four representative downstream tasks, covering text-video retrieval, video question answering, video action recognition, and temporal action localization.

We introduce VideoPrism, a general-purpose video encoder that tackles diverse video understanding tasks with a single frozen model. We pretrain VideoPrism on a heterogeneous corpus containing 36M high-quality video-caption pairs and 582M video clips with noisy parallel text (e.g., ASR transcripts). The pretraining approach improves upon masked autoencoding by global-local distillation of semantic video embeddings and a token shuffling scheme, enabling VideoPrism to focus primarily on the video modality while leveraging the invaluable text associated with videos. We extensively test VideoPrism on four broad groups of video understanding tasks, from web video question answering to CV for science, achieving state-of-the-art performance on 30 out of 33 video understanding benchmarks.

Core to understanding emotion are subjective experiences and their expression in facial behavior. Past studies have largely focused on six emotions and prototypical facial poses, reflecting limitations in scale and narrow assumptions about the variety of emotions and their patterns of expression. We examine 45,231 facial reactions to 2,185 evocative videos, largely in North America, Europe, and Japan, collecting participants’ self-reported experiences in English or Japanese and manual and automated annotations of facial movement. Guided by Semantic Space Theory, we uncover 21 dimensions of emotion in the self-reported experiences of participants in Japan, the United States, and Western Europe, and considerable cross-cultural similarities in experience. Facial expressions predict at least 12 dimensions of experience, despite massive individual differences in experience. We find considerable cross-cultural convergence in the facial actions involved in the expression of emotion, and culture-specific display tendencies—many facial movements differ in intensity in Japan compared to the U.S./Canada and Europe but represent similar experiences. These results quantitatively detail that people in dramatically different cultures experience and express emotion in a high-dimensional, categorical, and similar but complex fashion.

The recent advance in vision-language models is largely attributed to the abundance of image-text data. We aim to replicate this success for video-language models, but there simply is not enough human-curated video-text data available. We thus resort to fine-tuning a video-language model from a strong image-language baseline with syn-thesized instructional data. The resulting video model by video-instruction-tuning (VIIT) is then used to auto-label millions of videos to generate high-quality captions. We show the adapted video-language model performs well on a wide range of video-language benchmarks. For instance, it surpasses the best prior result on open-ended NExT-QA by 2.8%. Besides, our model generates detailed descriptions for previously unseen videos, which provide better textual supervision than existing methods. Experiments show that a video-language dual-encoder model contrastively trained on these auto-generated captions is 3.8% better than the strongest baseline that also leverages vision-language models. Our best model outperforms state-of-the-art methods on MSR-VTT zero-shot text-to-video retrieval by 6%. As a side product, we generate the largest video capation dataset to date.

This work focuses on training a single visual relationship detector predicting over the union of label spaces from multiple datasets. Merging labels spanning different datasets could be challenging due to inconsistent taxonomies. The issue is exacerbated in visual relationship detection when second-order visual semantics are introduced between pairs of objects. To address this challenge, we propose UniVRD, a novel bottom-up method for Unified Visual Relationship Detection by leveraging vision and language models (VLMs). VLMs provide well-aligned image and text embeddings, where similar relationships are optimized to be close to each other for semantic unification. Our bottom-up design enables the model to enjoy the benefit of training with both object detection and visual relationship datasets. Empirical results on both human-object interaction detection and scene-graph generation demonstrate the competitive performance of our model. UniVRD achieves 38.07 mAP on HICO-DET, outperforming the current best bottom-up HOI detector by 14.26 mAP. More importantly, we show that our unified detector performs as well as dataset-specific models in mAP, and achieves further improvements when we scale up the model. Our code will be made publicly available on GitHub.

Existing video-language pre-training methods primarily focus on instance-level alignment between video clips and captions via global contrastive learning but neglect rich fine-grained local information in both videos and text, which is of importance to downstream tasks requiring temporal localization and semantic reasoning. A powerful model is expected to be capable of capturing region-object correspondences and recognizing scene changes in a video clip, reflecting spatial and temporal granularity, respectively. To strengthen model's understanding into such fine-grained details, we propose a simple yet effective video-language modeling framework, S-ViLM, by exploiting the intrinsic structures of these two modalities. It includes two novel designs, inter-clip spatial grounding and intra-clip temporal grouping, to promote learning region-object alignment and temporal-aware features, simultaneously. Comprehensive evaluations demonstrate that S-ViLM performs favorably against existing approaches in learning more expressive representations. Specifically, S-ViLM surpasses the state-of-the-art methods substantially on four representative downstream tasks, covering text-video retrieval, video question answering, video action recognition, and temporal action localization.

Small on-device models have been successfully trained with user-level differential privacy (DP) for next word prediction and image classification tasks in the past. However, existing methods can fail when directly applied to learn embedding models using supervised training data with a large class space. To achieve user-level DP for large image-to-embedding feature extractors, we propose DP-FedEmb, a variant of federated learning algorithms with per-user sensitivity control and noise addition, to train from user-partitioned data centralized in the datacenter. DP-FedEmb combines virtual clients, partial aggregation, private local fine-tuning, and public Pre-training to achieve strong privacy utility trade-offs. We apply DP-FedEmb to train image embedding models for faces, landmarks and natural species, and demonstrate its superior utility under same privacy budget on benchmark datasets DigiFace. EMNIST, GLD and iNaturalist. We further illustrate it is possible to achieve strong user-level DP guarantees of ∊ < 2 while controlling the utility drop within 5%, when millions of users can participate in training.

Modern self-supervised learning algorithms typically enforce persistency of instance representations across views. While being very effective on learning holistic image and video representations, such an objective becomes suboptimal for learning spatio-temporally fine-grained features in videos, where scenes and instances evolve through space and time. In this paper, we present Contextualized Spatio-Temporal Contrastive Learning (ConST-CL) to effectively learn spatio-temporally fine-grained video representations via self-supervision. We first design a region-based pretext task which requires the model to transform instance representations from one view to another, guided by context features. Further, we introduce a simple network design that successfully reconciles the simultaneous learning process of both holistic and local representations. We evaluate our learned representations on a variety of downstream tasks and show that ConST-CL achieves competitive results on 6 datasets, including Kinetics, UCF, HMDB, AVA-Kinetics, AVA and OTB. Our code and models will be available at https://github.com/tensorflow/models/tree/master/official/projects/const_cl.

DeepLab2 is a TensorFlow library for deep labeling, aiming to provide a state-of-the-art and easy-to-use TensorFlow codebase for general dense pixel prediction problems in computer vision. DeepLab2 includes all our recently developed DeepLab model variants with pretrained checkpoints as well as model training and evaluation code, allowing the community to reproduce and further improve upon the state-of-art systems. To showcase the effectiveness of DeepLab2, our Panoptic-DeepLab employing Axial-SWideRNet as network backbone achieves 68.0% PQ or 83.5% mIoU on Cityscaspes validation set, with only single-scale inference and ImageNet-1K pretrained checkpoints. We hope that publicly sharing our library could facilitate future research on dense pixel labeling tasks and envision new applications of this technology. Code is made publicly available at \url{https://github.com/google-research/deeplab2}.

Recognition of human poses and actions is crucial for autonomous systems to interact smoothly with people. However, cameras generally capture human poses in 2D as images and videos, which can have significant appearance variations across viewpoints that make the recognition tasks challenging. To address this, we explore recognizing similarity in 3D human body poses from 2D information, which has not been well-studied in existing works. Here, we propose an approach to learning a compact view-invariant embedding space from 2D body joint keypoints, without explicitly predicting 3D poses. Input ambiguities of 2D poses from projection and occlusion are difficult to represent through a deterministic mapping, and therefore we adopt a probabilistic formulation for our embedding space. Experimental results show that our embedding model achieves higher accuracy when retrieving similar poses across different camera views, in comparison with 3D pose estimation models. We also show that by training a simple temporal embedding model, we achieve superior performance on pose sequence retrieval and largely reduce the embedding dimension from stacking frame-based embeddings for efficient large-scale retrieval. Furthermore, in order to enable our embeddings to work with partially visible input, we further investigate different keypoint occlusion augmentation strategies during training. We demonstrate that these occlusion augmentations significantly improve retrieval performance on partial 2D input poses. Results on action recognition and video alignment demonstrate that using our embeddings without any additional training achieves competitive performance relative to other models specifically trained for each task.

Core to understanding emotion are subjective experiences and their embodiment in facial behavior. Past studies have focused on six emotions and prototypical facial poses, reflecting limitations in scale and narrow assumptions about emotion. We examine 45,231 reactions to 2,185 evocative videos, largely in North America, Europe, and Japan, collecting participants’ self-reported experiences in English or Japanese and manual/automated annotations of facial movement. We uncover 21 dimensions of emotion underlying experiences reported across languages. Facial expressions predict at least 12 dimensions of experience, despite individual variability. We also identify culture-specific display tendencies—many facial movements differ in intensity in Japan compared to the U.S./Canada and Europe, but represent similar experiences. These results reveal how people actually experience and express emotion: in high-dimensional, categorical, and complex fashion.

We introduce a novel representation learning method to disentangle pose-dependent as well as view-dependent factors from 2D human poses. The method trains a network using cross-view mutual information maximization (CV-MIM) which maximizes mutual information of the same pose performed from different viewpoints in a contrastive learning manner. We further propose two regularization terms to ensure disentanglement and smoothness of the learned representations. The resulting pose representations can be used for cross-view action recognition. To evaluate the power of the learned representations, in addition to the conventional fully-supervised action recognition settings, we introduce a novel task called single-shot cross-view action recognition. This task trains models with actions from only one single viewpoint while models are evaluated on poses captured from all possible viewpoints. We evaluate the learned representations on standard benchmarks for action recognition, and show that (i) CV-MIM performs competitively compared with the state-of-the-art models in the fully-supervised scenarios; (ii) CV-MIM outperforms other competing methods by a large margin in the single-shot cross-view setting; (iii) and the learned representations can significantly boost the performance when reducing the amount of supervised training data. Our code is made publicly available at https://github.com/google-research/google-research/tree/master/poem

In this document, we cover the details of the implementation and experiments for our work. We also provide additional ablation studies and analysis. Specifically, we have: • Section 1 describes how we decide the NP-MPJPE threshold based on its effect on visual pose similarity. • Section 2 provides additional implementation details on model training, keypoint definition and normalization, downstream task experiment setup, etc. • Section 3 provides additional ablation studies, including the effect of key hyperparameters, ordered embedding variance visualizations, and embedding space visualization. • Section 4 provides additional quantitative pose retrieval result comparisons with image-based EpipolarPose model [5] for view-invariant pose retrieval. • Section 5 provides additional qualitative pose retrieval results. • Section 6 describes the qualitative video alignment experiment. Please refer to https://drive.google.com/open?id=1kTc_UT0Eq0H2ZBgfEoh8qEJMFBouC-Wv for the video synchronization results.

When we watch videos, the visual and auditory information we experience can evoke a range of affective responses. The ability to automatically predict evoked affect from videos can help recommendation systems and social machines better interact with their users. Here, we introduce the Evoked Expressions in Videos (EEV) dataset, a large-scale dataset for studying viewer responses to videos based on their facial expressions. The dataset consists of a total of 4.8 million annotations of viewer facial reactions to 18,541 videos. We use a publicly available video corpus to obtain a diverse set of video content. The training split is fully machine-annotated, while the validation and test splits have both human and machine annotations. We verify the performance of our machine annotations with human raters to have an average precision of 73.3%. We establish baseline performance on the EEV dataset using an existing multimodal recurrent model. Our results show that affective information can be learned from EEV, but with a MAP of 20.32%, there is potential for improvement. This gap motivates the need for new approaches for understanding affective content. Our transfer learning experiments show an improvement in performance on the LIRIS-ACCEDE video dataset when pre-trained on EEV. We hope that the size and diversity of the EEV dataset will encourage further explorations in video understanding and affective computing.

Understanding the degree to which human facial expressions co-vary with specific social contexts across cultures is central to the theory that emotions enable adaptive responses to important challenges and opportunities 1 – 6 . Concrete evidence linking social context to specific facial expressions is sparse and is largely based on survey-based approaches, which are often constrained by language and small sample sizes 7 – 13 . Here, by applying machine-learning methods to real-world, dynamic behaviour, we ascertain whether naturalistic social contexts (for example, weddings or sporting competitions) are associated with specific facial expressions 14 across different cultures. In two experiments using deep neural networks, we examined the extent to which 16 types of facial expression occurred systematically in thousands of contexts in 6 million videos from 144 countries. We found that each kind of facial expression had distinct associations with a set of contexts that were 70% preserved across 12 world regions. Consistent with these associations, regions varied in how frequently different facial expressions were produced as a function of which contexts were most salient. Our results reveal fine-grained patterns in human facial expressions that are preserved across the modern world.

Depictions of similar human body configurations can vary with changing viewpoints. Using only 2D information, we would like to enable vision algorithms to recognize similarity in human body poses across multiple views. This ability is useful for analyzing body movements and human behaviors in images and videos. In this paper, we propose an approach for learning a compact view-invariant embedding space from 2D joint keypoints alone, without explicitly predicting 3D poses. Since 2D poses are projected from 3D space, they have an inherent ambiguity, which is difficult to represent through a deterministic mapping. Hence, we use probabilistic embeddings to model this input uncertainty. Experimental results show that our embedding model achieves higher accuracy when retrieving similar poses across different camera views, in comparison with 2D-to-3D pose lifting models. We also demonstrate the effectiveness of applying our embeddings to view-invariant action recognition and video alignment. Our code is available at https://github.com/google-research/google-research/tree/master/poem .