We present CLIP2Video network to transfer the image-language pre-training model to video-text retrieval in an end-to-end manner. Leading approaches in the domain of video-and-language learning try to distill the spatio-temporal video features and multi-modal interaction between videos and languages from a large-scale video-text dataset. Different from them, we leverage pretrained image-language model, simplify it as a two-stage framework with co-learning of image-text and enhancing temporal relations between video frames and video-text respectively, make it able to train on comparatively small datasets. Specifically, based on the spatial semantics captured by Contrastive Language-Image Pretraining (CLIP) model, our model involves a Temporal Difference Block to capture motions at fine temporal video frames, and a Temporal Alignment Block to re-align the tokens of video clips and phrases and enhance the multi-modal correlation. We conduct thorough ablation studies, and achieve state-of-the-art performance on major text-to-video and video-to-text retrieval benchmarks, including new records of retrieval accuracy on MSR-VTT, MSVD and VATEX.
Instance-level contrastive learning techniques, which rely on data augmentation and a contrastive loss function, have found great success in the domain of visual representation learning. They are not suitable for exploiting the rich dynamical structure of video however, as operations are done on many augmented instances. In this paper we propose "Video Cross-Stream Prototypical Contrasting", a novel method which predicts consistent prototype assignments from both RGB and optical flow views, operating on sets of samples. Specifically, we alternate the optimization process; while optimizing one of the streams, all views are mapped to one set of stream prototype vectors. Each of the assignments is predicted with all views except the one matching the prediction, pushing representations closer to their assigned prototypes. As a result, more efficient video embeddings with ingrained motion information are learned, without the explicit need for optical flow computation during inference. We obtain state-of-the-art results on nearest-neighbour video retrieval and action recognition, outperforming previous best by +3.2% on UCF101 using the S3D backbone (90.5% Top-1 acc), and by +7.2% on UCF101 and +15.1% on HMDB51 using the R(2+1)D backbone.
Most existing video-and-language (VidL) research focuses on a single dataset, or multiple datasets of a single task. In reality, a truly useful VidL system is expected to be easily generalizable to diverse tasks, domains, and datasets. To facilitate the evaluation of such systems, we introduce Video-And-Language Understanding Evaluation (VALUE) benchmark, an assemblage of 11 VidL datasets over 3 popular tasks: (i) text-to-video retrieval; (ii) video question answering; and (iii) video captioning. VALUE benchmark aims to cover a broad range of video genres, video lengths, data volumes, and task difficulty levels. Rather than focusing on single-channel videos with visual information only, VALUE promotes models that leverage information from both video frames and their associated subtitles, as well as models that share knowledge across multiple tasks. We evaluate various baseline methods with and without large-scale VidL pre-training, and systematically investigate the impact of video input channels, fusion methods, and different video representations. We also study the transferability between tasks, and conduct multi-task learning under different settings. The significant gap between our best model and human performance calls for future study for advanced VidL models. VALUE is available at https://value-benchmark.github.io/.
We present a simplified, task-agnostic multi-modal pre-training approach that can accept either video or text input, or both for a variety of end tasks. Existing pre-training are task-specific by adopting either a single cross-modal encoder that requires both modalities, limiting their use for retrieval-style end tasks or more complex multitask learning with two unimodal encoders, limiting early cross-modal fusion. We instead introduce new pretraining masking schemes that better mix across modalities (e.g. by forcing masks for text to predict the closest video embeddings) while also maintaining separability (e.g. unimodal predictions are sometimes required, without using all the input). Experimental results show strong performance across a wider range of tasks than any previous methods, often outperforming task-specific pre-training. Code is made available at https://github.com/pytorch/fairseq/tree/main/examples/MMPT.
Multimodal self-supervised learning is getting more and more attention as it allows not only to train large networks without human supervision but also to search and retrieve data across various modalities. In this context, this paper proposes a self-supervised training framework that learns a common multimodal embedding space that, in addition to sharing representations across different modalities, enforces a grouping of semantically similar instances. To this end, we extend the concept of instance-level contrastive learning with a multimodal clustering step in the training pipeline to capture semantic similarities across modalities. The resulting embedding space enables retrieval of samples across all modalities, even from unseen datasets and different domains. To evaluate our approach, we train our model on the HowTo100M dataset and evaluate its zero-shot retrieval capabilities in two challenging domains, namely text-to-video retrieval, and temporal action localization, showing state-of-the-art results on four different datasets.
We present a framework for learning multimodal representations from unlabeled data using convolution-free Transformer architectures. Specifically, our Video-Audio-Text Transformer (VATT) takes raw signals as inputs and extracts multimodal representations that are rich enough to benefit a variety of downstream tasks. We train VATT end-to-end from scratch using multimodal contrastive losses and evaluate its performance by the downstream tasks of video action recognition, audio event classification, image classification, and text-to-video retrieval. Furthermore, we study a modality-agnostic, single-backbone Transformer by sharing weights among the three modalities. We show that the convolution-free VATT outperforms state-of-the-art ConvNet-based architectures in the downstream tasks. Especially, VATT's vision Transformer achieves the top-1 accuracy of 82.1% on Kinetics-400, 83.6% on Kinetics-600, 72.7% on Kinetics-700, and 41.1% on Moments in Time, new records while avoiding supervised pre-training. Transferring to image classification leads to 78.7% top-1 accuracy on ImageNet compared to 64.7% by training the same Transformer from scratch, showing the generalizability of our model despite the domain gap between videos and images. VATT's audio Transformer also sets a new record on waveform-based audio event recognition by achieving the mAP of 39.4% on AudioSet without any supervised pre-training. VATT's source code is publicly available.
Video-text retrieval plays an essential role in multi-modal research and has been widely used in many real-world web applications. The CLIP (Contrastive Language-Image Pre-training), an image-language pre-training model, has demonstrated the power of visual concepts learning from web collected image-text datasets. In this paper, we propose a CLIP4Clip model to transfer the knowledge of the CLIP model to video-language retrieval in an end-to-end manner. Several questions are investigated via empirical studies: 1) Whether image feature is enough for video-text retrieval? 2) How a post-pretraining on a large-scale video-text dataset based on the CLIP affect the performance? 3) What is the practical mechanism to model temporal dependency between video frames? And 4) The Hyper-parameters sensitivity of the model on video-text retrieval task. Extensive experimental results present that the CLIP4Clip model transferred from the CLIP can achieve SOTA results on various video-text retrieval datasets, including MSR-VTT, MSVC, LSMDC, ActivityNet, and DiDeMo. We release our code at https://github.com/ArrowLuo/CLIP4Clip.
In recent years, considerable progress on the task of text-video retrieval has been achieved by leveraging large-scale pretraining on visual and audio datasets to construct powerful video encoders. By contrast, despite the natural symmetry, the design of effective algorithms for exploiting large-scale language pretraining remains under-explored. In this work, we are the first to investigate the design of such algorithms and propose a novel generalized distillation method, TeachText, which leverages complementary cues from multiple text encoders to provide an enhanced supervisory signal to the retrieval model. Moreover, we extend our method to video side modalities and show that we can effectively reduce the number of used modalities at test time without compromising performance. Our approach advances the state of the art on several video retrieval benchmarks by a significant margin and adds no computational overhead at test time. Last but not least, we show an effective application of our method for eliminating noise from retrieval datasets. Code and data can be found at https://www.robots.ox.ac.uk/~vgg/research/teachtext/.
This work strives for the classification and localization of human actions in videos, without the need for any labeled video training examples. Where existing work relies on transferring global attribute or object information from seen to unseen action videos, we seek to classify and spatio-temporally localize unseen actions in videos from image-based object information only. We propose three spatial object priors, which encode local person and object detectors along with their spatial relations. On top we introduce three semantic object priors, which extend semantic matching through word embeddings with three simple functions that tackle semantic ambiguity, object discrimination, and object naming. A video embedding combines the spatial and semantic object priors. It enables us to introduce a new video retrieval task that retrieves action tubes in video collections based on user-specified objects, spatial relations, and object size. Experimental evaluation on five action datasets shows the importance of spatial and semantic object priors for unseen actions. We find that persons and objects have preferred spatial relations that benefit unseen action localization, while using multiple languages and simple object filtering directly improves semantic matching, leading to state-of-the-art results for both unseen action classification and localization.
A key challenge in self-supervised video representation learning is how to effectively capture motion information besides context bias. While most existing works implicitly achieve this with video-specific pretext tasks (e.g., predicting clip orders, time arrows, and paces), we develop a method that explicitly decouples motion supervision from context bias through a carefully designed pretext task. Specifically, we take the keyframes and motion vectors in compressed videos (e.g., in H.264 format) as the supervision sources for context and motion, respectively, which can be efficiently extracted at over 500 fps on the CPU. Then we design two pretext tasks that are jointly optimized: a context matching task where a pairwise contrastive loss is cast between video clip and keyframe features; and a motion prediction task where clip features, passed through an encoder-decoder network, are used to estimate motion features in a near future. These two tasks use a shared video backbone and separate MLP heads. Experiments show that our approach improves the quality of the learned video representation over previous works, where we obtain absolute gains of 16.0% and 11.1% in video retrieval recall on UCF101 and HMDB51, respectively. Moreover, we find the motion prediction to be a strong regularization for video networks, where using it as an auxiliary task improves the accuracy of action recognition with a margin of 7.4%~13.8%.
Our objective in this work is video-text retrieval - in particular a joint embedding that enables efficient text-to-video retrieval. The challenges in this area include the design of the visual architecture and the nature of the training data, in that the available large scale video-text training datasets, such as HowTo100M, are noisy and hence competitive performance is achieved only at scale through large amounts of compute. We address both these challenges in this paper. We propose an end-to-end trainable model that is designed to take advantage of both large-scale image and video captioning datasets. Our model is an adaptation and extension of the recent ViT and Timesformer architectures, and consists of attention in both space and time. The model is flexible and can be trained on both image and video text datasets, either independently or in conjunction. It is trained with a curriculum learning schedule that begins by treating images as 'frozen' snapshots of video, and then gradually learns to attend to increasing temporal context when trained on video datasets. We also provide a new video-text pretraining dataset WebVid-2M, comprised of over two million videos with weak captions scraped from the internet. Despite training on datasets that are an order of magnitude smaller, we show that this approach yields state-of-the-art results on standard downstream video-retrieval benchmarks including MSR-VTT, MSVD, DiDeMo and LSMDC.
We present a new state-of-the-art on the text to video retrieval task on MSRVTT and LSMDC benchmarks where our model outperforms all previous solutions by a large margin. Moreover, state-of-the-art results are achieved with a single model on two datasets without finetuning. This multidomain generalisation is achieved by a proper combination of different video caption datasets. We show that training on different datasets can improve test results of each other. Additionally we check intersection between many popular datasets and found that MSRVTT has a significant overlap between the test and the train parts, and the same situation is observed for ActivityNet.
Current video retrieval efforts all found their evaluation on an instance-based assumption, that only a single caption is relevant to a query video and vice versa. We demonstrate that this assumption results in performance comparisons often not indicative of models' retrieval capabilities. We propose a move to semantic similarity video retrieval, where (i) multiple videos/captions can be deemed equally relevant, and their relative ranking does not affect a method's reported performance and (ii) retrieved videos/captions are ranked by their similarity to a query. We propose several proxies to estimate semantic similarities in large-scale retrieval datasets, without additional annotations. Our analysis is performed on three commonly used video retrieval datasets (MSR-VTT, YouCook2 and EPIC-KITCHENS).
Video retrieval using natural language queries requires learning semantically meaningful joint embeddings between the text and the audio-visual input. Often, such joint embeddings are learnt using pairwise (or triplet) contrastive loss objectives which cannot give enough attention to 'difficult-to-retrieve' samples during training. This problem is especially pronounced in data-scarce settings where the data is relatively small (10% of the large scale MSR-VTT) to cover the rather complex audio-visual embedding space. In this context, we introduce Rudder - a multilingual video-text retrieval dataset that includes audio and textual captions in Marathi, Hindi, Tamil, Kannada, Malayalam and Telugu. Furthermore, we propose to compensate for data scarcity by using domain knowledge to augment supervision. To this end, in addition to the conventional three samples of a triplet (anchor, positive, and negative), we introduce a fourth term - a partial - to define a differential margin based partialorder loss. The partials are heuristically sampled such that they semantically lie in the overlap zone between the positives and the negatives, thereby resulting in broader embedding coverage. Our proposals consistently outperform the conventional max-margin and triplet losses and improve the state-of-the-art on MSR-VTT and DiDeMO datasets. We report benchmark results on Rudder while also observing significant gains using the proposed partial order loss, especially when the language specific retrieval models are jointly trained by availing the cross-lingual alignment across the language-specific datasets.
Video Retrieval is a challenging task where a text query is matched to a video or vice versa. Most of the existing approaches for addressing such a problem rely on annotations made by the users. Although simple, this approach is not always feasible in practice. In this work, we explore the application of the language-image model, CLIP, to obtain video representations without the need for said annotations. This model was explicitly trained to learn a common space where images and text can be compared. Using various techniques described in this document, we extended its application to videos, obtaining state-of-the-art results on the MSR-VTT and MSVD benchmarks.
Visual Place Recognition (VPR) is the task of matching current visual imagery from a camera to images stored in a reference map of the environment. While initial VPR systems used simple direct image methods or hand-crafted visual features, recent work has focused on learning more powerful visual features and further improving performance through either some form of sequential matcher / filter or a hierarchical matching process. In both cases the performance of the initial single-image based system is still far from perfect, putting significant pressure on the sequence matching or (in the case of hierarchical systems) pose refinement stages. In this paper we present a novel hybrid system that creates a high performance initial match hypothesis generator using short learnt sequential descriptors, which enable selective control sequential score aggregation using single image learnt descriptors. Sequential descriptors are generated using a temporal convolutional network dubbed SeqNet, encoding short image sequences using 1-D convolutions, which are then matched against the corresponding temporal descriptors from the reference dataset to provide an ordered list of place match hypotheses. We then perform selective sequential score aggregation using shortlisted single image learnt descriptors from a separate pipeline to produce an overall place match hypothesis. Comprehensive experiments on challenging benchmark datasets demonstrate the proposed method outperforming recent state-of-the-art methods using the same amount of sequential information. Source code and supplementary material can be found at https://github.com/oravus/seqNet.
Current video/action understanding systems have demonstrated impressive performance on large recognition tasks. However, they might be limiting themselves to learning to recognize spatiotemporal patterns, rather than attempting to thoroughly understand the actions. To spur progress in the direction of a truer, deeper understanding of videos, we introduce the task of win-fail action recognition -- differentiating between successful and failed attempts at various activities. We introduce a first of its kind paired win-fail action understanding dataset with samples from the following domains: "General Stunts," "Internet Wins-Fails," "Trick Shots," and "Party Games." Unlike existing action recognition datasets, intra-class variation is high making the task challenging, yet feasible. We systematically analyze the characteristics of the win-fail task/dataset with prototypical action recognition networks and a novel video retrieval task. While current action recognition methods work well on our task/dataset, they still leave a large gap to achieve high performance. We hope to motivate more work towards the true understanding of actions/videos. Dataset will be available from https://github.com/ParitoshParmar/Win-Fail-Action-Recognition.
The canonical approach to video-and-language learning (e.g., video question answering) dictates a neural model to learn from offline-extracted dense video features from vision models and text features from language models. These feature extractors are trained independently and usually on tasks different from the target domains, rendering these fixed features sub-optimal for downstream tasks. Moreover, due to the high computational overload of dense video features, it is often difficult (or infeasible) to plug feature extractors directly into existing approaches for easy finetuning. To provide a remedy to this dilemma, we propose a generic framework ClipBERT that enables affordable end-to-end learning for video-and-language tasks, by employing sparse sampling, where only a single or a few sparsely sampled short clips from a video are used at each training step. Experiments on text-to-video retrieval and video question answering on six datasets demonstrate that ClipBERT outperforms (or is on par with) existing methods that exploit full-length videos, suggesting that end-to-end learning with just a few sparsely sampled clips is often more accurate than using densely extracted offline features from full-length videos, proving the proverbial less-is-more principle. Videos in the datasets are from considerably different domains and lengths, ranging from 3-second generic domain GIF videos to 180-second YouTube human activity videos, showing the generalization ability of our approach. Comprehensive ablation studies and thorough analyses are provided to dissect what factors lead to this success. Our code is publicly available at https://github.com/jayleicn/ClipBERT
Retrieving unlabeled videos by textual queries, known as Ad-hoc Video Search (AVS), is a core theme in multimedia data management and retrieval. The success of AVS counts on cross-modal representation learning that encodes both query sentences and videos into common spaces for semantic similarity computation. Inspired by the initial success of previously few works in combining multiple sentence encoders, this paper takes a step forward by developing a new and general method for effectively exploiting diverse sentence encoders. The novelty of the proposed method, which we term Sentence Encoder Assembly (SEA), is two-fold. First, different from prior art that use only a single common space, SEA supports text-video matching in multiple encoder-specific common spaces. Such a property prevents the matching from being dominated by a specific encoder that produces an encoding vector much longer than other encoders. Second, in order to explore complementarities among the individual common spaces, we propose multi-space multi-loss learning. As extensive experiments on four benchmarks (MSR-VTT, TRECVID AVS 2016-2019, TGIF and MSVD) show, SEA surpasses the state-of-the-art. In addition, SEA is extremely ease to implement. All this makes SEA an appealing solution for AVS and promising for continuously advancing the task by harvesting new sentence encoders.
We study unsupervised video representation learning that seeks to learn both motion and appearance features from unlabeled video only, which can be reused for downstream tasks such as action recognition. This task, however, is extremely challenging due to 1) the highly complex spatial-temporal information in videos; and 2) the lack of labeled data for training. Unlike the representation learning for static images, it is difficult to construct a suitable self-supervised task to well model both motion and appearance features. More recently, several attempts have been made to learn video representation through video playback speed prediction. However, it is non-trivial to obtain precise speed labels for the videos. More critically, the learnt models may tend to focus on motion pattern and thus may not learn appearance features well. In this paper, we observe that the relative playback speed is more consistent with motion pattern, and thus provide more effective and stable supervision for representation learning. Therefore, we propose a new way to perceive the playback speed and exploit the relative speed between two video clips as labels. In this way, we are able to well perceive speed and learn better motion features. Moreover, to ensure the learning of appearance features, we further propose an appearance-focused task, where we enforce the model to perceive the appearance difference between two video clips. We show that optimizing the two tasks jointly consistently improves the performance on two downstream tasks, namely action recognition and video retrieval. Remarkably, for action recognition on UCF101 dataset, we achieve 93.7% accuracy without the use of labeled data for pre-training, which outperforms the ImageNet supervised pre-trained model. Code and pre-trained models can be found at https://github.com/PeihaoChen/RSPNet.