Most multimodal multi-objective evolutionary algorithms (MMEAs) aim to find all global Pareto optimal sets (PSs) for a multimodal multi-objective optimization problem (MMOP). However, in real-world problems, decision makers (DMs) may be also interested in local PSs. Also, searching for both global and local PSs is more general in view of dealing with MMOPs, which can be seen as a generalized MMOP. In addition, the state-of-the-art MMEAs exhibit poor convergence on high-dimension MMOPs. To address the above two issues, in this study, a novel coevolutionary framework termed CoMMEA for multimodal multi-objective optimization is proposed to better obtain both global and local PSs, and simultaneously, to improve the convergence performance in dealing with high-dimension MMOPs. Specifically, the CoMMEA introduces two archives to the search process, and coevolves them simultaneously through effective knowledge transfer. The convergence archive assists the CoMMEA to quickly approaching the Pareto optimal front (PF). The knowledge of the converged solutions is then transferred to the diversity archive which utilizes the local convergence indicator and the $\epsilon$-dominance-based method to obtain global and local PSs effectively. Experimental results show that CoMMEA is competitive compared to seven state-of-the-art MMEAs on fifty-four complex MMOPs.
Error correction in automatic speech recognition (ASR) aims to correct those incorrect words in sentences generated by ASR models. Since recent ASR models usually have low word error rate (WER), to avoid affecting originally correct tokens, error correction models should only modify incorrect words, and therefore detecting incorrect words is important for error correction. Previous works on error correction either implicitly detect error words through target-source attention or CTC (connectionist temporal classification) loss, or explicitly locate specific deletion/substitution/insertion errors. However, implicit error detection does not provide clear signal about which tokens are incorrect and explicit error detection suffers from low detection accuracy. In this paper, we propose SoftCorrect with a soft error detection mechanism to avoid the limitations of both explicit and implicit error detection. Specifically, we first detect whether a token is correct or not through a probability produced by a dedicatedly designed language model, and then design a constrained CTC loss that only duplicates the detected incorrect tokens to let the decoder focus on the correction of error tokens. Compared with implicit error detection with CTC loss, SoftCorrect provides explicit signal about which words are incorrect and thus does not need to duplicate every token but only incorrect tokens; compared with explicit error detection, SoftCorrect does not detect specific deletion/substitution/insertion errors but just leaves it to CTC loss. Experiments on AISHELL-1 and Aidatatang datasets show that SoftCorrect achieves 26.1% and 9.4% CER reduction respectively, outperforming previous works by a large margin, while still enjoying fast speed of parallel generation.
Place recognition based on point cloud (LiDAR) scans is an important module for achieving robust autonomy in robots or self-driving vehicles. Training deep networks to match such scans presents a difficult trade-off: a higher spatial resolution of the network's intermediate representations is needed to perform fine-grained matching of subtle geometric features, but growing it too large makes the memory requirements infeasible. In this work, we propose a Point-Voxel Transformer network (PVT3D) that achieves robust fine-grained matching with low memory requirements. It leverages a sparse voxel branch to extract and aggregate information at a lower resolution and a point-wise branch to obtain fine-grained local information. A novel hierarchical cross-attention transformer (HCAT) uses queries from one branch to try to match structures in the other branch, ensuring that both extract self-contained descriptors of the point cloud (rather than one branch dominating), but using both to inform the output global descriptor of the point cloud. Extensive experiments show that the proposed PVT3D method surpasses the state-of-the-art by a large amount on several datasets (Oxford RobotCar, TUM, USyd). For instance, we achieve AR@1 of 85.6% on the TUM dataset, which surpasses the strongest prior model by ~15%.
Objection detection (OD) has been one of the most fundamental tasks in computer vision. Recent developments in deep learning have pushed the performance of image OD to new heights by learning-based, data-driven approaches. On the other hand, video OD remains less explored, mostly due to much more expensive data annotation needs. At the same time, multi-object tracking (MOT) which requires reasoning about track identities and spatio-temporal trajectories, shares similar spirits with video OD. However, most MOT datasets are class-specific (e.g., person-annotated only), which constrains a model's flexibility to perform tracking on other objects. We propose TrIVD (Tracking and Image-Video Detection), the first framework that unifies image OD, video OD, and MOT within one end-to-end model. To handle the discrepancies and semantic overlaps across datasets, TrIVD formulates detection/tracking as grounding and reasons about object categories via visual-text alignments. The unified formulation enables cross-dataset, multi-task training, and thus equips TrIVD with the ability to leverage frame-level features, video-level spatio-temporal relations, as well as track identity associations. With such joint training, we can now extend the knowledge from OD data, that comes with much richer object category annotations, to MOT and achieve zero-shot tracking capability. Experiments demonstrate that TrIVD achieves state-of-the-art performances across all image/video OD and MOT tasks.
Video-based gait recognition has achieved impressive results in constrained scenarios. However, visual cameras neglect human 3D structure information, which limits the feasibility of gait recognition in the 3D wild world. In this work, instead of extracting gait features from images, we explore precise 3D gait features from point clouds and propose a simple yet efficient 3D gait recognition framework, termed multi-view projection network (MVPNet). MVPNet first projects point clouds into multiple depth maps from different perspectives, and then fuse depth images together, to learn the compact representation with 3D geometry information. Due to the lack of point cloud datasets, we build the first large-scale Lidar-based gait recognition dataset, LIDAR GAIT, collected by a Lidar sensor and an RGB camera mounted on a robot. The dataset contains 25,279 sequences from 1,050 subjects and covers many different variations, including visibility, views, occlusions, clothing, carrying, and scenes. Extensive experiments show that, (1) 3D structure information serves as a significant feature for gait recognition. (2) MVPNet not only competes with five representative point-based methods, but it also outperforms existing camera-based methods by large margins. (3) The Lidar sensor is superior to the RGB camera for gait recognition in the wild. LIDAR GAIT dataset and MVPNet code will be publicly available.
Indoor scenes typically exhibit complex, spatially-varying appearance from global illumination, making inverse rendering a challenging ill-posed problem. This work presents an end-to-end, learning-based inverse rendering framework incorporating differentiable Monte Carlo raytracing with importance sampling. The framework takes a single image as input to jointly recover the underlying geometry, spatially-varying lighting, and photorealistic materials. Specifically, we introduce a physically-based differentiable rendering layer with screen-space ray tracing, resulting in more realistic specular reflections that match the input photo. In addition, we create a large-scale, photorealistic indoor scene dataset with significantly richer details like complex furniture and dedicated decorations. Further, we design a novel out-of-view lighting network with uncertainty-aware refinement leveraging hypernetwork-based neural radiance fields to predict lighting outside the view of the input photo. Through extensive evaluations on common benchmark datasets, we demonstrate superior inverse rendering quality of our method compared to state-of-the-art baselines, enabling various applications such as complex object insertion and material editing with high fidelity. Code and data will be made available at \url{https://jingsenzhu.github.io/invrend}.
In this paper, we introduce the task of learning unsupervised dialogue embeddings. Trivial approaches such as combining pre-trained word or sentence embeddings and encoding through pre-trained language models (PLMs) have been shown to be feasible for this task. However, these approaches typically ignore the conversational interactions between interlocutors, resulting in poor performance. To address this issue, we proposed a self-guided contrastive learning approach named dial2vec. Dial2vec considers a dialogue as an information exchange process. It captures the conversational interaction patterns between interlocutors and leverages them to guide the learning of the embeddings corresponding to each interlocutor. The dialogue embedding is obtained by an aggregation of the embeddings from all interlocutors. To verify our approach, we establish a comprehensive benchmark consisting of six widely-used dialogue datasets. We consider three evaluation tasks: domain categorization, semantic relatedness, and dialogue retrieval. Dial2vec achieves on average 8.7, 9.0, and 13.8 points absolute improvements in terms of purity, Spearman's correlation, and mean average precision (MAP) over the strongest baseline on the three tasks respectively. Further analysis shows that dial2vec obtains informative and discriminative embeddings for both interlocutors under the guidance of the conversational interactions and achieves the best performance when aggregating them through the interlocutor-level pooling strategy. All codes and data are publicly available at https://github.com/AlibabaResearch/DAMO-ConvAI/tree/main/dial2vec.
Symbolic music generation aims to generate music scores automatically. A recent trend is to use Transformer or its variants in music generation, which is, however, suboptimal, because the full attention cannot efficiently model the typically long music sequences (e.g., over 10,000 tokens), and the existing models have shortcomings in generating musical repetition structures. In this paper, we propose Museformer, a Transformer with a novel fine- and coarse-grained attention for music generation. Specifically, with the fine-grained attention, a token of a specific bar directly attends to all the tokens of the bars that are most relevant to music structures (e.g., the previous 1st, 2nd, 4th and 8th bars, selected via similarity statistics); with the coarse-grained attention, a token only attends to the summarization of the other bars rather than each token of them so as to reduce the computational cost. The advantages are two-fold. First, it can capture both music structure-related correlations via the fine-grained attention, and other contextual information via the coarse-grained attention. Second, it is efficient and can model over 3X longer music sequences compared to its full-attention counterpart. Both objective and subjective experimental results demonstrate its ability to generate long music sequences with high quality and better structures.
Edge intelligence autonomous driving (EIAD) offers computing resources in autonomous vehicles for training deep neural networks. However, wireless channels between the edge server and autonomous vehicles are time-varying due to the high-mobility of vehicles. Moreover, the required number of training samples for different data modalities (e.g., images, point-clouds) is diverse. Consequently, when collecting these datasets from vehicles to the edge server, the associated bandwidth and power allocation across all data frames is a large-scale multi-modal optimization problem. This article proposes a highly computationally efficient algorithm that directly maximizes the quality of training (QoT). The key ingredients include a data-driven model for quantifying the priority of data modality and two first-order methods termed accelerated gradient projection and dual decomposition for low-complexity resource allocation. High-fidelity simulations in Car Learning to Act (CARLA) show that the proposed algorithm reduces the perception error by $3\%$ and the computation time by $98\%$.
This paper describes Tencent AI Lab - Shanghai Jiao Tong University (TAL-SJTU) Low-Resource Translation systems for the WMT22 shared task. We participate in the general translation task on English$\Leftrightarrow$Livonian. Our system is based on M2M100 with novel techniques that adapt it to the target language pair. (1) Cross-model word embedding alignment: inspired by cross-lingual word embedding alignment, we successfully transfer a pre-trained word embedding to M2M100, enabling it to support Livonian. (2) Gradual adaptation strategy: we exploit Estonian and Latvian as auxiliary languages for many-to-many translation training and then adapt to English-Livonian. (3) Data augmentation: to enlarge the parallel data for English-Livonian, we construct pseudo-parallel data with Estonian and Latvian as pivot languages. (4) Fine-tuning: to make the most of all available data, we fine-tune the model with the validation set and online back-translation, further boosting the performance. In model evaluation: (1) We find that previous work underestimated the translation performance of Livonian due to inconsistent Unicode normalization, which may cause a discrepancy of up to 14.9 BLEU score. (2) In addition to the standard validation set, we also employ round-trip BLEU to evaluate the models, which we find more appropriate for this task. Finally, our unconstrained system achieves BLEU scores of 17.0 and 30.4 for English to/from Livonian.