3D object detection plays a crucial role in numerous intelligent vision systems. Detection in the open world inevitably encounters various adverse scenes, such as dense fog, heavy rain, and low light conditions. Although existing efforts primarily focus on diversifying network architecture or training schemes, resulting in significant progress in 3D object detection, most of these learnable modules fail in adverse scenes, thereby hindering detection performance. To address this issue, this paper proposes a monocular 3D detection model designed to perceive twin depth in adverse scenes, termed MonoTDP, which effectively mitigates the degradation of detection performance in various harsh environments. Specifically, we first introduce an adaptive learning strategy to aid the model in handling uncontrollable weather conditions, significantly resisting degradation caused by various degrading factors. Then, to address the depth/content loss in adverse regions, we propose a novel twin depth perception module that simultaneously estimates scene and object depth, enabling the integration of scene-level features and object-level features. Additionally, we assemble a new adverse 3D object detection dataset encompassing a wide range of challenging scenes, including rainy, foggy, and low light weather conditions, with each type of scene containing 7,481 images. Experimental results demonstrate that our proposed method outperforms current state-of-the-art approaches by an average of 3.12% in terms of AP_R40 for car category across various adverse environments.
Recently, multi-modality scene perception tasks, e.g., image fusion and scene understanding, have attracted widespread attention for intelligent vision systems. However, early efforts always consider boosting a single task unilaterally and neglecting others, seldom investigating their underlying connections for joint promotion. To overcome these limitations, we establish the hierarchical dual tasks-driven deep model to bridge these tasks. Concretely, we firstly construct an image fusion module to fuse complementary characteristics and cascade dual task-related modules, including a discriminator for visual effects and a semantic network for feature measurement. We provide a bi-level perspective to formulate image fusion and follow-up downstream tasks. To incorporate distinct task-related responses for image fusion, we consider image fusion as a primary goal and dual modules as learnable constraints. Furthermore, we develop an efficient first-order approximation to compute corresponding gradients and present dynamic weighted aggregation to balance the gradients for fusion learning. Extensive experiments demonstrate the superiority of our method, which not only produces visually pleasant fused results but also realizes significant promotion for detection and segmentation than the state-of-the-art approaches.
The MultiCoNER II task aims to detect complex, ambiguous, and fine-grained named entities in low-context situations and noisy scenarios like the presence of spelling mistakes and typos for multiple languages. The task poses significant challenges due to the scarcity of contextual information, the high granularity of the entities(up to 33 classes), and the interference of noisy data. To address these issues, our team {\bf PAI} proposes a universal Named Entity Recognition (NER) system that integrates external entity information to improve performance. Specifically, our system retrieves entities with properties from the knowledge base (i.e. Wikipedia) for a given text, then concatenates entity information with the input sentence and feeds it into Transformer-based models. Finally, our system wins 2 first places, 4 second places, and 1 third place out of 13 tracks. The code is publicly available at \url{https://github.com/diqiuzhuanzhuan/semeval-2023}.
Wav2vec 2.0 (W2V2) has shown impressive performance in automatic speech recognition (ASR). However, the large model size and the non-streaming architecture make it hard to be used under low-resource or streaming scenarios. In this work, we propose a two-stage knowledge distillation method to solve these two problems: the first step is to make the big and non-streaming teacher model smaller, and the second step is to make it streaming. Specially, we adopt the MSE loss for the distillation of hidden layers and the modified LF-MMI loss for the distillation of the prediction layer. Experiments are conducted on Gigaspeech, Librispeech, and an in-house dataset. The results show that the distilled student model (DistillW2V2) we finally get is 8x faster and 12x smaller than the original teacher model. For the 480ms latency setup, the DistillW2V2's relative word error rate (WER) degradation varies from 9% to 23.4% on test sets, which reveals a promising way to extend the W2V2's application scope.
It is difficult for an end-to-end (E2E) ASR system to recognize words such as named entities appearing infrequently in the training data. A widely used method to mitigate this issue is feeding contextual information into the acoustic model. A contextual word list is necessary, which lists all possible contextual word candidates. Previous works have proven that the size and quality of the list are crucial. A compact and accurate list can boost the performance significantly. In this paper, we propose an efficient approach to obtain a high quality contextual word list for a unified streaming and non-streaming based Conformer-Transducer (C-T) model. Specifically, we make use of the phone-level streaming output to first filter the predefined contextual word list. During the subsequent non-streaming inference, the words in the filtered list are regarded as contextual information fused into non-casual encoder and decoder to generate the final recognition results. Our approach can take advantage of streaming recognition hypothesis, improve the accuracy of the contextual ASR system and speed up the inference process as well. Experiments on two datasets demonstrates over 20% relative character error rate reduction (CERR) comparing to the baseline system. Meanwile, the RTF of our system can be stabilized within 0.15 when the size of the contextual word list grows over 6,000.
Improving the visual quality of the given degraded observation by correcting exposure level is a fundamental task in the computer vision community. Existing works commonly lack adaptability towards unknown scenes because of the data-driven patterns (deep networks) and limited regularization (traditional optimization), and they usually need time-consuming inference. These two points heavily limit their practicability. In this paper, we establish a Practical Exposure Corrector (PEC) that assembles the characteristics of efficiency and performance. To be concrete, we rethink the exposure correction to provide a linear solution with exposure-sensitive compensation. Around generating the compensation, we introduce an exposure adversarial function as the key engine to fully extract valuable information from the observation. By applying the defined function, we construct a segmented shrinkage iterative scheme to generate the desired compensation. Its shrinkage nature supplies powerful support for algorithmic stability and robustness. Extensive experimental evaluations fully reveal the superiority of our proposed PEC. The code is available at https://rsliu.tech/PEC.
Underwater image enhancement has become an attractive topic as a significant technology in marine engineering and aquatic robotics. However, the limited number of datasets and imperfect hand-crafted ground truth weaken its robustness to unseen scenarios, and hamper the application to high-level vision tasks. To address the above limitations, we develop an efficient and compact enhancement network in collaboration with a high-level semantic-aware pretrained model, aiming to exploit its hierarchical feature representation as an auxiliary for the low-level underwater image enhancement. Specifically, we tend to characterize the shallow layer features as textures while the deep layer features as structures in the semantic-aware model, and propose a multi-path Contextual Feature Refinement Module (CFRM) to refine features in multiple scales and model the correlation between different features. In addition, a feature dominative network is devised to perform channel-wise modulation on the aggregated texture and structure features for the adaptation to different feature patterns of the enhancement network. Extensive experiments on benchmarks demonstrate that the proposed algorithm achieves more appealing results and outperforms state-of-the-art methods by large margins. We also apply the proposed algorithm to the underwater salient object detection task to reveal the favorable semantic-aware ability for high-level vision tasks. The code is available at STSC.
More and more attention has been paid to the segmentation of pulmonary nodules. Among the current methods based on deep learning, 3D segmentation methods directly input 3D images, which takes up a lot of memory and brings huge computation. However, most of the 2D segmentation methods with less parameters and calculation have the problem of lacking spatial relations between slices, resulting in poor segmentation performance. In order to solve these problems, we propose an adjacent slice feature guided 2.5D network. In this paper, we design an adjacent slice feature fusion model to introduce information from adjacent slices. To further improve the model performance, we construct a multi-scale fusion module to capture more context information, in addition, we design an edge-constrained loss function to optimize the segmentation results in the edge region. Fully experiments show that our method performs better than other existing methods in pulmonary nodule segmentation task.
Customized keyword spotting (KWS) has great potential to be deployed on edge devices to achieve hands-free user experience. However, in real applications, false alarm (FA) would be a serious problem for spotting dozens or even hundreds of keywords, which drastically affects user experience. To solve this problem, in this paper, we leverage the recent advances in transducer and transformer based acoustic models and propose a new multi-stage customized KWS framework named Cascaded Transducer-Transformer KWS (CaTT-KWS), which includes a transducer based keyword detector, a frame-level phone predictor based force alignment module and a transformer based decoder. Specifically, the streaming transducer module is used to spot keyword candidates in audio stream. Then force alignment is implemented using the phone posteriors predicted by the phone predictor to finish the first stage keyword verification and refine the time boundaries of keyword. Finally, the transformer decoder further verifies the triggered keyword. Our proposed CaTT-KWS framework reduces FA rate effectively without obviously hurting keyword recognition accuracy. Specifically, we can get impressively 0.13 FA per hour on a challenging dataset, with over 90% relative reduction on FA comparing to the transducer based detection model, while keyword recognition accuracy only drops less than 2%.
Leveraging context information is an intuitive idea to improve performance on conversational automatic speech recognition(ASR). Previous works usually adopt recognized hypotheses of historical utterances as preceding context, which may bias the current recognized hypothesis due to the inevitable historicalrecognition errors. To avoid this problem, we propose an audio-textual cross-modal representation extractor to learn contextual representations directly from preceding speech. Specifically, it consists of two modal-related encoders, extracting high-level latent features from speech and the corresponding text, and a cross-modal encoder, which aims to learn the correlation between speech and text. We randomly mask some input tokens and input sequences of each modality. Then a token-missing or modal-missing prediction with a modal-level CTC loss on the cross-modal encoder is performed. Thus, the model captures not only the bi-directional context dependencies in a specific modality but also relationships between different modalities. Then, during the training of the conversational ASR system, the extractor will be frozen to extract the textual representation of preceding speech, while such representation is used as context fed to the ASR decoder through attention mechanism. The effectiveness of the proposed approach is validated on several Mandarin conversation corpora and the highest character error rate (CER) reduction up to 16% is achieved on the MagicData dataset.