Learning from videos is an emerging research area that enables robots to acquire skills from human demonstrations, such as procedural videos. To do this, video-language models must be able to obtain structured understandings, such as the temporal segmentation of a demonstration into sequences of actions and skills, and to generalize the understandings to novel domains. In pursuit of this goal, we introduce Spacewalk-18, a benchmark containing two tasks: (1) step recognition and (2) intra-video retrieval over a dataset of temporally segmented and labeled tasks in International Space Station spacewalk recordings. In tandem, the two tasks quantify a model's ability to make use of: (1) out-of-domain visual information; (2) a high temporal context window; and (3) multimodal (e.g. visual and speech) domains. This departs from existing benchmarks for procedural video understanding, which typically deal with short context lengths and can be solved with a single modality. Spacewalk-18, with its inherent multimodal and long-form complexity, exposes the high difficulty of task recognition and segmentation. We find that state-of-the-art methods perform poorly on our benchmark, but improvements can be obtained by incorporating information from longer-range temporal context across different modalities. Our experiments underscore the need to develop new approaches to these tasks. Data, model, and code will be released at https://brown-palm.github.io/Spacewalk-18/.
To build scalable models for challenging real-world tasks, it is important to learn from diverse, multi-modal data in various forms (e.g., videos, text, and images). Among the existing works, a plethora of them have focused on leveraging large but cumbersome cross-modal architectures. Regardless of their effectiveness, larger architectures unavoidably prevent the models from being extended to real-world applications, so building a lightweight VL architecture and an efficient learning schema is of great practical value. In this paper, we propose an Efficient Video-Language Model (dubbed as E-ViLM) and a masked video modeling (MVM) schema, assisted with a semantic vector-quantized tokenizer. In particular, our E-ViLM learns to reconstruct the semantic labels of masked video regions, produced by the pre-trained vector-quantized tokenizer, which discretizes the continuous visual signals into labels. We show that with our simple MVM task and regular VL pre-training modelings, our E-ViLM, despite its compactness, is able to learn expressive representations from Video-Language corpus and generalize well to extensive Video-Language tasks including video question answering, text-to-video retrieval, etc. In particular, our E-ViLM obtains obvious efficiency improvements by reaching competing performances with faster inference speed, i.e., our model reaches $39.3$% Top-$1$ accuracy on the MSRVTT benchmark, retaining $91.4$% of the accuracy of state-of-the-art larger VL architecture with only $15%$ parameters and $94.8%$ fewer GFLOPs. We also provide extensive ablative studies that validate the effectiveness of our proposed learning schema for E-ViLM.
Large pre-trained vision models achieve impressive success in computer vision. However, fully fine-tuning large models for downstream tasks, particularly in video understanding, can be prohibitively computationally expensive. Recent studies turn their focus towards efficient image-to-video transfer learning. Nevertheless, existing efficient fine-tuning methods lack attention to training memory usage and exploration of transferring a larger model to the video domain. In this paper, we present a novel Spatial-Temporal Side Network for memory-efficient fine-tuning large image models to video understanding, named Side4Video. Specifically, we introduce a lightweight spatial-temporal side network attached to the frozen vision model, which avoids the backpropagation through the heavy pre-trained model and utilizes multi-level spatial features from the original image model. Extremely memory-efficient architecture enables our method to reduce 75% memory usage than previous adapter-based methods. In this way, we can transfer a huge ViT-E (4.4B) for video understanding tasks which is 14x larger than ViT-L (304M). Our approach achieves remarkable performance on various video datasets across unimodal and cross-modal tasks (i.e., action recognition and text-video retrieval), especially in Something-Something V1&V2 (67.3% & 74.6%), Kinetics-400 (88.6%), MSR-VTT (52.3%), MSVD (56.1%) and VATEX (68.8%). We release our code at https://github.com/HJYao00/Side4Video.
Despite being (pre)trained on a massive amount of data, state-of-the-art video-language alignment models are not robust to semantically-plausible contrastive changes in the video captions. Our work addresses this by identifying a broad spectrum of contrast misalignments, such as replacing entities, actions, and flipping event order, which alignment models should be robust against. To this end, we introduce the VideoCon, a video-language alignment dataset constructed by a large language model that generates plausible contrast video captions and explanations for differences between original and contrast video captions. Then, a generative video-language model is finetuned with VideoCon to assess video-language entailment and generate explanations. Our VideoCon-based alignment model significantly outperforms current models. It exhibits a 12-point increase in AUC for the video-language alignment task on human-generated contrast captions. Finally, our model sets new state of the art zero-shot performance in temporally-extensive video-language tasks such as text-to-video retrieval (SSv2-Temporal) and video question answering (ATP-Hard). Moreover, our model shows superior performance on novel videos and human-crafted captions and explanations. Our code and data are available at https://github.com/Hritikbansal/videocon.
A recent trend in multimodal retrieval is related to postprocessing test set results via the dual-softmax loss (DSL). While this approach can bring significant improvements, it usually presumes that an entire matrix of test samples is available as DSL input. This work introduces a new postprocessing approach based on Sinkhorn transformations that outperforms DSL. Further, we propose a new postprocessing setting that does not require access to multiple test queries. We show that our approach can significantly improve the results of state of the art models such as CLIP4Clip, BLIP, X-CLIP, and DRL, thus achieving a new state-of-the-art on several standard text-video retrieval datasets both with access to the entire test set and in the single-query setting.
Fashion stylists have historically bridged the gap between consumers' desires and perfect outfits, which involve intricate combinations of colors, patterns, and materials. Although recent advancements in fashion recommendation systems have made strides in outfit compatibility prediction and complementary item retrieval, these systems rely heavily on pre-selected customer choices. Therefore, we introduce a groundbreaking approach to fashion recommendations: text-to-outfit retrieval task that generates a complete outfit set based solely on textual descriptions given by users. Our model is devised at three semantic levels-item, style, and outfit-where each level progressively aggregates data to form a coherent outfit recommendation based on textual input. Here, we leverage strategies similar to those in the contrastive language-image pretraining model to address the intricate-style matrix within the outfit sets. Using the Maryland Polyvore and Polyvore Outfit datasets, our approach significantly outperformed state-of-the-art models in text-video retrieval tasks, solidifying its effectiveness in the fashion recommendation domain. This research not only pioneers a new facet of fashion recommendation systems, but also introduces a method that captures the essence of individual style preferences through textual descriptions.
Text-to-video retrieval (TVR) aims to find the most relevant video in a large video gallery given a query text. The intricate and abundant context of the video challenges the performance and efficiency of TVR. To handle the serialized video contexts, existing methods typically select a subset of frames within a video to represent the video content for TVR. How to select the most representative frames is a crucial issue, whereby the selected frames are required to not only retain the semantic information of the video but also promote retrieval efficiency by excluding temporally redundant frames. In this paper, we make the first empirical study of frame selection for TVR. We systemically classify existing frame selection methods into text-free and text-guided ones, under which we detailedly analyze six different frame selections in terms of effectiveness and efficiency. Among them, two frame selections are first developed in this paper. According to the comprehensive analysis on multiple TVR benchmarks, we empirically conclude that the TVR with proper frame selections can significantly improve the retrieval efficiency without sacrificing the retrieval performance.
Large-scale video-language pre-training has made remarkable strides in advancing video-language understanding tasks. However, the heavy computational burden of video encoding remains a formidable efficiency bottleneck, particularly for long-form videos. These videos contain massive visual tokens due to their inherent 3D properties and spatiotemporal redundancy, making it challenging to capture complex temporal and spatial relationships. To tackle this issue, we propose an efficient method called TEmporal-Spatial Token Aggregation (TESTA). TESTA condenses video semantics by adaptively aggregating similar frames, as well as similar patches within each frame. TESTA can reduce the number of visual tokens by 75% and thus accelerate video encoding. Building upon TESTA, we introduce a pre-trained video-language model equipped with a divided space-time token aggregation module in each video encoder block. We evaluate our model on five datasets for paragraph-to-video retrieval and long-form VideoQA tasks. Experimental results show that TESTA improves computing efficiency by 1.7 times, and achieves significant performance gains from its scalability in processing longer input frames, e.g., +13.7 R@1 on QuerYD and +6.5 R@1 on Condensed Movie.
Compressing videos into binary codes can improve retrieval speed and reduce storage overhead. However, learning accurate hash codes for video retrieval can be challenging due to high local redundancy and complex global dependencies between video frames, especially in the absence of labels. Existing self-supervised video hashing methods have been effective in designing expressive temporal encoders, but have not fully utilized the temporal dynamics and spatial appearance of videos due to less challenging and unreliable learning tasks. To address these challenges, we begin by utilizing the contrastive learning task to capture global spatio-temporal information of videos for hashing. With the aid of our designed augmentation strategies, which focus on spatial and temporal variations to create positive pairs, the learning framework can generate hash codes that are invariant to motion, scale, and viewpoint. Furthermore, we incorporate two collaborative learning tasks, i.e., frame order verification and scene change regularization, to capture local spatio-temporal details within video frames, thereby enhancing the perception of temporal structure and the modeling of spatio-temporal relationships. Our proposed Contrastive Hashing with Global-Local Spatio-temporal Information (CHAIN) outperforms state-of-the-art self-supervised video hashing methods on four video benchmark datasets. Our codes will be released.
Vision-language models (VLMs) classify the query video by calculating a similarity score between the visual features and text-based class label representations. Recently, large language models (LLMs) have been used to enrich the text-based class labels by enhancing the descriptiveness of the class names. However, these improvements are restricted to the text-based classifier only, and the query visual features are not considered. In this paper, we propose a framework which combines pre-trained discriminative VLMs with pre-trained generative video-to-text and text-to-text models. We introduce two key modifications to the standard zero-shot setting. First, we propose language-guided visual feature enhancement and employ a video-to-text model to convert the query video to its descriptive form. The resulting descriptions contain vital visual cues of the query video, such as what objects are present and their spatio-temporal interactions. These descriptive cues provide additional semantic knowledge to VLMs to enhance their zeroshot performance. Second, we propose video-specific prompts to LLMs to generate more meaningful descriptions to enrich class label representations. Specifically, we introduce prompt techniques to create a Tree Hierarchy of Categories for class names, offering a higher-level action context for additional visual cues, We demonstrate the effectiveness of our approach in video understanding across three different zero-shot settings: 1) video action recognition, 2) video-to-text and textto-video retrieval, and 3) time-sensitive video tasks. Consistent improvements across multiple benchmarks and with various VLMs demonstrate the effectiveness of our proposed framework. Our code will be made publicly available.
Unsupervised video hashing usually optimizes binary codes by learning to reconstruct input videos. Such reconstruction constraint spends much effort on frame-level temporal context changes without focusing on video-level global semantics that are more useful for retrieval. Hence, we address this problem by decomposing video information into reconstruction-dependent and semantic-dependent information, which disentangles the semantic extraction from reconstruction constraint. Specifically, we first design a simple dual-stream structure, including a temporal layer and a hash layer. Then, with the help of semantic similarity knowledge obtained from self-supervision, the hash layer learns to capture information for semantic retrieval, while the temporal layer learns to capture the information for reconstruction. In this way, the model naturally preserves the disentangled semantics into binary codes. Validated by comprehensive experiments, our method consistently outperforms the state-of-the-arts on three video benchmarks.
Despite significant results achieved by Contrastive Language-Image Pretraining (CLIP) in zero-shot image recognition, limited effort has been made exploring its potential for zero-shot video recognition. This paper presents Open-VCLIP++, a simple yet effective framework that adapts CLIP to a strong zero-shot video classifier, capable of identifying novel actions and events during testing. Open-VCLIP++ minimally modifies CLIP to capture spatial-temporal relationships in videos, thereby creating a specialized video classifier while striving for generalization. We formally demonstrate that training Open-VCLIP++ is tantamount to continual learning with zero historical data. To address this problem, we introduce Interpolated Weight Optimization, a technique that leverages the advantages of weight interpolation during both training and testing. Furthermore, we build upon large language models to produce fine-grained video descriptions. These detailed descriptions are further aligned with video features, facilitating a better transfer of CLIP to the video domain. Our approach is evaluated on three widely used action recognition datasets, following a variety of zero-shot evaluation protocols. The results demonstrate that our method surpasses existing state-of-the-art techniques by significant margins. Specifically, we achieve zero-shot accuracy scores of 88.1%, 58.7%, and 81.2% on UCF, HMDB, and Kinetics-600 datasets respectively, outpacing the best-performing alternative methods by 8.5%, 8.2%, and 12.3%. We also evaluate our approach on the MSR-VTT video-text retrieval dataset, where it delivers competitive video-to-text and text-to-video retrieval performance, while utilizing substantially less fine-tuning data compared to other methods. Code is released at https://github.com/wengzejia1/Open-VCLIP.
Given a text query, partially relevant video retrieval (PRVR) seeks to find untrimmed videos containing pertinent moments in a database. For PRVR, clip modeling is essential to capture the partial relationship between texts and videos. Current PRVR methods adopt scanning-based clip construction to achieve explicit clip modeling, which is information-redundant and requires a large storage overhead. To solve the efficiency problem of PRVR methods, this paper proposes GMMFormer, a Gaussian-Mixture-Model based Transformer which models clip representations implicitly. During frame interactions, we incorporate Gaussian-Mixture-Model constraints to focus each frame on its adjacent frames instead of the whole video. Then generated representations will contain multi-scale clip information, achieving implicit clip modeling. In addition, PRVR methods ignore semantic differences between text queries relevant to the same video, leading to a sparse embedding space. We propose a query diverse loss to distinguish these text queries, making the embedding space more intensive and contain more semantic information. Extensive experiments on three large-scale video datasets (i.e., TVR, ActivityNet Captions, and Charades-STA) demonstrate the superiority and efficiency of GMMFormer. Code is available at \url{https://github.com/huangmozhi9527/GMMFormer}.
Foundational multimodal models pre-trained on large scale image-text pairs or video-text pairs or both have shown strong generalization abilities on downstream tasks. However unlike image-text models, pretraining video-text models is always not feasible due to the difficulty in collecting large-scale clean and aligned data, and exponential computational costs involved in the pretraining phase. Therefore, the pertinent question to ask is: Can image-text models be adapted to video tasks and is there any benefit to using these models over pretraining directly on videos? In this work, we focus on this question by proposing a detailed study on the generalization abilities of image-text models when evaluated on video understanding tasks in a zero-shot setting. We investigate 9 foundational image-text models on a diverse set of video tasks that include video action recognition (video AR), video retrieval (video RT), video question answering (video QA), video multiple choice (video MC) and video captioning (video CP). Our experiments show that image-text models exhibit impressive performance on video AR, video RT and video MC. Furthermore, they perform moderately on video captioning and poorly on video QA. These findings shed a light on the benefits of adapting foundational image-text models to an array of video tasks while avoiding the costly pretraining step.
Instructional videos are a common source for learning text-video or even multimodal representations by leveraging subtitles extracted with automatic speech recognition systems (ASR) from the audio signal in the videos. However, in contrast to human-annotated captions, both speech and subtitles naturally differ from the visual content of the videos and thus provide only noisy supervision. As a result, large-scale annotation-free web video training data remains sub-optimal for training text-video models. In this work, we propose to leverage the capabilities of large language models (LLMs) to obtain high-quality video descriptions aligned with videos at scale. Specifically, we prompt an LLM to create plausible video captions based on ASR subtitles of instructional videos. To this end, we introduce a prompting method that is able to take into account a longer text of subtitles, allowing us to capture the contextual information beyond one single sentence. We further prompt the LLM to generate timestamps for each produced caption based on the timestamps of the subtitles and finally align the generated captions to the video temporally. In this way, we obtain human-style video captions at scale without human supervision. We apply our method to the subtitles of the HowTo100M dataset, creating a new large-scale dataset, HowToCaption. Our evaluation shows that the resulting captions not only significantly improve the performance over many different benchmark datasets for zero-shot text-video retrieval and video captioning, but also lead to a disentangling of textual narration from the audio, boosting the performance in text-video-audio tasks.
With the explosive growth of web videos in recent years, large-scale Content-Based Video Retrieval (CBVR) becomes increasingly essential in video filtering, recommendation, and copyright protection. Segment-level CBVR (S-CBVR) locates the start and end time of similar segments in finer granularity, which is beneficial for user browsing efficiency and infringement detection especially in long video scenarios. The challenge of S-CBVR task is how to achieve high temporal alignment accuracy with efficient computation and low storage consumption. In this paper, we propose a Segment Similarity and Alignment Network (SSAN) in dealing with the challenge which is firstly trained end-to-end in S-CBVR. SSAN is based on two newly proposed modules in video retrieval: (1) An efficient Self-supervised Keyframe Extraction (SKE) module to reduce redundant frame features, (2) A robust Similarity Pattern Detection (SPD) module for temporal alignment. In comparison with uniform frame extraction, SKE not only saves feature storage and search time, but also introduces comparable accuracy and limited extra computation time. In terms of temporal alignment, SPD localizes similar segments with higher accuracy and efficiency than existing deep learning methods. Furthermore, we jointly train SSAN with SKE and SPD and achieve an end-to-end improvement. Meanwhile, the two key modules SKE and SPD can also be effectively inserted into other video retrieval pipelines and gain considerable performance improvements. Experimental results on public datasets show that SSAN can obtain higher alignment accuracy while saving storage and online query computational cost compared to existing methods.
In recent years, the explosion of web videos makes text-video retrieval increasingly essential and popular for video filtering, recommendation, and search. Text-video retrieval aims to rank relevant text/video higher than irrelevant ones. The core of this task is to precisely measure the cross-modal similarity between texts and videos. Recently, contrastive learning methods have shown promising results for text-video retrieval, most of which focus on the construction of positive and negative pairs to learn text and video representations. Nevertheless, they do not pay enough attention to hard negative pairs and lack the ability to model different levels of semantic similarity. To address these two issues, this paper improves contrastive learning using two novel techniques. First, to exploit hard examples for robust discriminative power, we propose a novel Dual-Modal Attention-Enhanced Module (DMAE) to mine hard negative pairs from textual and visual clues. By further introducing a Negative-aware InfoNCE (NegNCE) loss, we are able to adaptively identify all these hard negatives and explicitly highlight their impacts in the training loss. Second, our work argues that triplet samples can better model fine-grained semantic similarity compared to pairwise samples. We thereby present a new Triplet Partial Margin Contrastive Learning (TPM-CL) module to construct partial order triplet samples by automatically generating fine-grained hard negatives for matched text-video pairs. The proposed TPM-CL designs an adaptive token masking strategy with cross-modal interaction to model subtle semantic differences. Extensive experiments demonstrate that the proposed approach outperforms existing methods on four widely-used text-video retrieval datasets, including MSR-VTT, MSVD, DiDeMo and ActivityNet.
The canonical approach to video-text retrieval leverages a coarse-grained or fine-grained alignment between visual and textual information. However, retrieving the correct video according to the text query is often challenging as it requires the ability to reason about both high-level (scene) and low-level (object) visual clues and how they relate to the text query. To this end, we propose a Unified Coarse-to-fine Alignment model, dubbed UCoFiA. Specifically, our model captures the cross-modal similarity information at different granularity levels. To alleviate the effect of irrelevant visual clues, we also apply an Interactive Similarity Aggregation module (ISA) to consider the importance of different visual features while aggregating the cross-modal similarity to obtain a similarity score for each granularity. Finally, we apply the Sinkhorn-Knopp algorithm to normalize the similarities of each level before summing them, alleviating over- and under-representation issues at different levels. By jointly considering the crossmodal similarity of different granularity, UCoFiA allows the effective unification of multi-grained alignments. Empirically, UCoFiA outperforms previous state-of-the-art CLIP-based methods on multiple video-text retrieval benchmarks, achieving 2.4%, 1.4% and 1.3% improvements in text-to-video retrieval R@1 on MSR-VTT, Activity-Net, and DiDeMo, respectively. Our code is publicly available at https://github.com/Ziyang412/UCoFiA.
Many studies focus on improving pretraining or developing new backbones in text-video retrieval. However, existing methods may suffer from the learning and inference bias issue, as recent research suggests in other text-video-related tasks. For instance, spatial appearance features on action recognition or temporal object co-occurrences on video scene graph generation could induce spurious correlations. In this work, we present a unique and systematic study of a temporal bias due to frame length discrepancy between training and test sets of trimmed video clips, which is the first such attempt for a text-video retrieval task, to the best of our knowledge. We first hypothesise and verify the bias on how it would affect the model illustrated with a baseline study. Then, we propose a causal debiasing approach and perform extensive experiments and ablation studies on the Epic-Kitchens-100, YouCook2, and MSR-VTT datasets. Our model overpasses the baseline and SOTA on nDCG, a semantic-relevancy-focused evaluation metric which proves the bias is mitigated, as well as on the other conventional metrics.
Large-scale noisy web image-text datasets have been proven to be efficient for learning robust vision-language models. However, when transferring them to the task of video retrieval, models still need to be fine-tuned on hand-curated paired text-video data to adapt to the diverse styles of video descriptions. To address this problem without the need for hand-annotated pairs, we propose a new setting, text-video retrieval with uncurated & unpaired data, that during training utilizes only text queries together with uncurated web videos without any paired text-video data. To this end, we propose an approach, In-Style, that learns the style of the text queries and transfers it to uncurated web videos. Moreover, to improve generalization, we show that one model can be trained with multiple text styles. To this end, we introduce a multi-style contrastive training procedure that improves the generalizability over several datasets simultaneously. We evaluate our model on retrieval performance over multiple datasets to demonstrate the advantages of our style transfer framework on the new task of uncurated & unpaired text-video retrieval and improve state-of-the-art performance on zero-shot text-video retrieval.