Few-shot (FS) and zero-shot (ZS) learning are two different approaches for scaling temporal action detection (TAD) to new classes. The former adapts a pretrained vision model to a new task represented by as few as a single video per class, whilst the latter requires no training examples by exploiting a semantic description of the new class. In this work, we introduce a new multi-modality few-shot (MMFS) TAD problem, which can be considered as a marriage of FS-TAD and ZS-TAD by leveraging few-shot support videos and new class names jointly. To tackle this problem, we further introduce a novel MUlti-modality PromPt mETa-learning (MUPPET) method. This is enabled by efficiently bridging pretrained vision and language models whilst maximally reusing already learned capacity. Concretely, we construct multi-modal prompts by mapping support videos into the textual token space of a vision-language model using a meta-learned adapter-equipped visual semantics tokenizer. To tackle large intra-class variation, we further design a query feature regulation scheme. Extensive experiments on ActivityNetv1.3 and THUMOS14 demonstrate that our MUPPET outperforms state-of-the-art alternative methods, often by a large margin. We also show that our MUPPET can be easily extended to tackle the few-shot object detection problem and again achieves the state-of-the-art performance on MS-COCO dataset. The code will be available in https://github.com/sauradip/MUPPET
Large-scale weakly supervised product retrieval is a practically useful yet computationally challenging problem. This paper introduces a novel solution for the eBay Visual Search Challenge (eProduct) held at the Ninth Workshop on Fine-Grained Visual Categorisation workshop (FGVC9) of CVPR 2022. This competition presents two challenges: (a) E-commerce is a drastically fine-grained domain including many products with subtle visual differences; (b) A lacking of target instance-level labels for model training, with only coarse category labels and product titles available. To overcome these obstacles, we formulate a strong solution by a set of dedicated designs: (a) Instead of using text training data directly, we mine thousands of pseudo-attributes from product titles and use them as the ground truths for multi-label classification. (b) We incorporate several strong backbones with advanced training recipes for more discriminative representation learning. (c) We further introduce a number of post-processing techniques including whitening, re-ranking and model ensemble for retrieval enhancement. By achieving 71.53% MAR, our solution "Involution King" achieves the second position on the leaderboard.
Existing temporal action detection (TAD) methods rely on large training data including segment-level annotations, limited to recognizing previously seen classes alone during inference. Collecting and annotating a large training set for each class of interest is costly and hence unscalable. Zero-shot TAD (ZS-TAD) resolves this obstacle by enabling a pre-trained model to recognize any unseen action classes. Meanwhile, ZS-TAD is also much more challenging with significantly less investigation. Inspired by the success of zero-shot image classification aided by vision-language (ViL) models such as CLIP, we aim to tackle the more complex TAD task. An intuitive method is to integrate an off-the-shelf proposal detector with CLIP style classification. However, due to the sequential localization (e.g, proposal generation) and classification design, it is prone to localization error propagation. To overcome this problem, in this paper we propose a novel zero-Shot Temporal Action detection model via Vision-LanguagE prompting (STALE). Such a novel design effectively eliminates the dependence between localization and classification by breaking the route for error propagation in-between. We further introduce an interaction mechanism between classification and localization for improved optimization. Extensive experiments on standard ZS-TAD video benchmarks show that our STALE significantly outperforms state-of-the-art alternatives. Besides, our model also yields superior results on supervised TAD over recent strong competitors. The PyTorch implementation of STALE is available at https://github.com/sauradip/STALE.
Existing temporal action detection (TAD) methods rely on a large number of training data with segment-level annotations. Collecting and annotating such a training set is thus highly expensive and unscalable. Semi-supervised TAD (SS-TAD) alleviates this problem by leveraging unlabeled videos freely available at scale. However, SS-TAD is also a much more challenging problem than supervised TAD, and consequently much under-studied. Prior SS-TAD methods directly combine an existing proposal-based TAD method and a SSL method. Due to their sequential localization (e.g, proposal generation) and classification design, they are prone to proposal error propagation. To overcome this limitation, in this work we propose a novel Semi-supervised Temporal action detection model based on PropOsal-free Temporal mask (SPOT) with a parallel localization (mask generation) and classification architecture. Such a novel design effectively eliminates the dependence between localization and classification by cutting off the route for error propagation in-between. We further introduce an interaction mechanism between classification and localization for prediction refinement, and a new pretext task for self-supervised model pre-training. Extensive experiments on two standard benchmarks show that our SPOT outperforms state-of-the-art alternatives, often by a large margin. The PyTorch implementation of SPOT is available at https://github.com/sauradip/SPOT
Existing temporal action detection (TAD) methods rely on generating an overwhelmingly large number of proposals per video. This leads to complex model designs due to proposal generation and/or per-proposal action instance evaluation and the resultant high computational cost. In this work, for the first time, we propose a proposal-free Temporal Action detection model with Global Segmentation mask (TAGS). Our core idea is to learn a global segmentation mask of each action instance jointly at the full video length. The TAGS model differs significantly from the conventional proposal-based methods by focusing on global temporal representation learning to directly detect local start and end points of action instances without proposals. Further, by modeling TAD holistically rather than locally at the individual proposal level, TAGS needs a much simpler model architecture with lower computational cost. Extensive experiments show that despite its simpler design, TAGS outperforms existing TAD methods, achieving new state-of-the-art performance on two benchmarks. Importantly, it is ~ 20x faster to train and ~1.6x more efficient for inference. Our PyTorch implementation of TAGS is available at https://github.com/sauradip/TAGS .
In this paper we present a novel self-supervised method to anticipate the depth estimate for a future, unobserved real-world urban scene. This work is the first to explore self-supervised learning for estimation of monocular depth of future unobserved frames of a video. Existing works rely on a large number of annotated samples to generate the probabilistic prediction of depth for unseen frames. However, this makes it unrealistic due to its requirement for large amount of annotated depth samples of video. In addition, the probabilistic nature of the case, where one past can have multiple future outcomes often leads to incorrect depth estimates. Unlike previous methods, we model the depth estimation of the unobserved frame as a view-synthesis problem, which treats the depth estimate of the unseen video frame as an auxiliary task while synthesizing back the views using learned pose. This approach is not only cost effective - we do not use any ground truth depth for training (hence practical) but also deterministic (a sequence of past frames map to an immediate future). To address this task we first develop a novel depth forecasting network DeFNet which estimates depth of unobserved future by forecasting latent features. Second, we develop a channel-attention based pose estimation network that estimates the pose of the unobserved frame. Using this learned pose, estimated depth map is reconstructed back into the image domain, thus forming a self-supervised solution. Our proposed approach shows significant improvements in Abs Rel metric compared to state-of-the-art alternatives on both short and mid-term forecasting setting, benchmarked on KITTI and Cityscapes. Code is available at https://github.com/sauradip/depthForecasting
Existing temporal action localization (TAL) works rely on a large number of training videos with exhaustive segment-level annotation, preventing them from scaling to new classes. As a solution to this problem, few-shot TAL (FS-TAL) aims to adapt a model to a new class represented by as few as a single video. Exiting FS-TAL methods assume trimmed training videos for new classes. However, this setting is not only unnatural actions are typically captured in untrimmed videos, but also ignores background video segments containing vital contextual cues for foreground action segmentation. In this work, we first propose a new FS-TAL setting by proposing to use untrimmed training videos. Further, a novel FS-TAL model is proposed which maximizes the knowledge transfer from training classes whilst enabling the model to be dynamically adapted to both the new class and each video of that class simultaneously. This is achieved by introducing a query adaptive Transformer in the model. Extensive experiments on two action localization benchmarks demonstrate that our method can outperform all the state of the art alternatives significantly in both single-domain and cross-domain scenarios. The source code can be found in https://github.com/sauradip/fewshotQAT
Detecting text located on the torsos of marathon runners and sports players in video is a challenging issue due to poor quality and adverse effects caused by flexible/colorful clothing, and different structures of human bodies or actions. This paper presents a new unified method for tackling the above challenges. The proposed method fuses gradient magnitude and direction coherence of text pixels in a new way for detecting candidate regions. Candidate regions are used for determining the number of temporal frame clusters obtained by K-means clustering on frame differences. This process in turn detects key frames. The proposed method explores Bayesian probability for skin portions using color values at both pixel and component levels of temporal frames, which provides fused images with skin components. Based on skin information, the proposed method then detects faces and torsos by finding structural and spatial coherences between them. We further propose adaptive pixels linking a deep learning model for text detection from torso regions. The proposed method is tested on our own dataset collected from marathon/sports video and three standard datasets, namely, RBNR, MMM and R-ID of marathon images, to evaluate the performance. In addition, the proposed method is also tested on the standard natural scene datasets, namely, CTW1500 and MS-COCO text datasets, to show the objectiveness of the proposed method. A comparative study with the state-of-the-art methods on bib number/text detection of different datasets shows that the proposed method outperforms the existing methods.
Salient object detection is a prevalent computer vision task that has applications ranging from abnormality detection to abnormality processing. Context modelling is an important criterion in the domain of saliency detection. A global context helps in determining the salient object in a given image by contrasting away other objects in the global view of the scene. However, the local context features detects the boundaries of the salient object with higher accuracy in a given region. To incorporate the best of both worlds, our proposed SaLite model uses both global and local contextual features. It is an encoder-decoder based architecture in which the encoder uses a lightweight SqueezeNet and decoder is modelled using convolution layers. Modern deep based models entitled for saliency detection use a large number of parameters, which is difficult to deploy on embedded systems. This paper attempts to solve the above problem using SaLite which is a lighter process for salient object detection without compromising on performance. Our approach is extensively evaluated on three publicly available datasets namely DUTS, MSRA10K, and SOC. Experimental results show that our proposed SaLite has significant and consistent improvements over the state-of-the-art methods.
Facial micro-expressions are sudden involuntary minute muscle movements which reveal true emotions that people try to conceal. Spotting a micro-expression and recognizing it is a major challenge owing to its short duration and intensity. Many works pursued traditional and deep learning based approaches to solve this issue but compromised on learning low-level features and higher accuracy due to unavailability of datasets. This motivated us to propose a novel joint architecture of spatial and temporal network which extracts time-contrasted features from the feature maps to contrast out micro-expression from rapid muscle movements. The usage of time contrasted features greatly improved the spotting of micro-expression from inconspicuous facial movements. Also, we include a memory module to predict the class and intensity of the micro-expression across the temporal frames of the micro-expression clip. Our method achieves superior performance in comparison to other conventional approaches on CASMEII dataset.