Since its introduction, the transformer model has demonstrated outstanding performance across various tasks. However, there are still unresolved issues regarding length generalization, particularly in algorithmic tasks. In this paper, we investigate the inherent capabilities of transformer models in learning arithmetic algorithms, such as addition and multiplication. Through experiments and attention analysis, we identify a number of crucial factors for achieving optimal length generalization. We show that transformer models are able to generalize to long lengths with the help of targeted attention biasing. We then introduce Attention Bias Calibration (ABC), a calibration stage that enables the model to automatically learn the proper attention biases, which we link to mechanisms in relative position encoding. We demonstrate that using ABC, the transformer model can achieve unprecedented perfect length generalization on certain arithmetic tasks.
Grid-centric perception is a crucial field for mobile robot perception and navigation. Nonetheless, grid-centric perception is less prevalent than object-centric perception for autonomous driving as autonomous vehicles need to accurately perceive highly dynamic, large-scale outdoor traffic scenarios and the complexity and computational costs of grid-centric perception are high. The rapid development of deep learning techniques and hardware gives fresh insights into the evolution of grid-centric perception and enables the deployment of many real-time algorithms. Current industrial and academic research demonstrates the great advantages of grid-centric perception, such as comprehensive fine-grained environmental representation, greater robustness to occlusion, more efficient sensor fusion, and safer planning policies. Given the lack of current surveys for this rapidly expanding field, we present a hierarchically-structured review of grid-centric perception for autonomous vehicles. We organize previous and current knowledge of occupancy grid techniques and provide a systematic in-depth analysis of algorithms in terms of three aspects: feature representation, data utility, and applications in autonomous driving systems. Lastly, we present a summary of the current research trend and provide some probable future outlooks.
Occupancy maps are widely recognized as an efficient method for facilitating robot motion planning in static environments. However, for intelligent vehicles, occupancy of both the present and future moments is required to ensure safe driving. In the automotive industry, the accurate and continuous prediction of future occupancy maps in traffic scenarios remains a formidable challenge. This paper investigates multi-sensor spatio-temporal fusion strategies for continuous occupancy prediction in a systematic manner. This paper presents FusionMotion, a novel bird's eye view (BEV) occupancy predictor which is capable of achieving the fusion of asynchronous multi-sensor data and predicting the future occupancy map with variable time intervals and temporal horizons. Remarkably, FusionMotion features the adoption of neural ordinary differential equations on recurrent neural networks for occupancy prediction. FusionMotion learns derivatives of BEV features over temporal horizons, updates the implicit sensor's BEV feature measurements and propagates future states for each ODE step. Extensive experiments on large-scale nuScenes and Lyft L5 datasets demonstrate that FusionMotion significantly outperforms previous methods. In addition, it outperforms the BEVFusion-style fusion strategy on the Lyft L5 dataset while reducing synchronization requirements. Codes and models will be made available.
The idea of cooperative perception is to benefit from shared perception data between multiple vehicles and overcome the limitations of on-board sensors on single vehicle. However, the fusion of multi-vehicle information is still challenging due to inaccurate localization, limited communication bandwidth and ambiguous fusion. Past practices simplify the problem by placing a precise GNSS localization system, manually specify the number of connected vehicles and determine the fusion strategy. This paper proposes a map-based cooperative perception framework, named map container, to improve the accuracy and robustness of cooperative perception, which ultimately overcomes this problem. The concept 'Map Container' denotes that the map serves as the platform to transform all information into the map coordinate space automatically and incorporate different sources of information in a distributed fusion architecture. In the proposed map container, the GNSS signal and the matching relationship between sensor feature and map feature are considered to optimize the estimation of environment states. Evaluation on simulation dataset and real-vehicle platform result validates the effectiveness of the proposed method.
Environmental perception with multi-modal fusion of radar and camera is crucial in autonomous driving to increase the accuracy, completeness, and robustness. This paper focuses on how to utilize millimeter-wave (MMW) radar and camera sensor fusion for 3D object detection. A novel method which realizes the feature-level fusion under bird-eye view (BEV) for a better feature representation is proposed. Firstly, radar features are augmented with temporal accumulation and sent to a temporal-spatial encoder for radar feature extraction. Meanwhile, multi-scale image 2D features which adapt to various spatial scales are obtained by image backbone and neck model. Then, image features are transformed to BEV with the designed view transformer. In addition, this work fuses the multi-modal features with a two-stage fusion model called point fusion and ROI fusion, respectively. Finally, a detection head regresses objects category and 3D locations. Experimental results demonstrate that the proposed method realizes the state-of-the-art performance under the most important detection metrics, mean average precision (mAP) and nuScenes detection score (NDS) on the challenging nuScenes dataset.
Detection And Tracking of Moving Objects (DATMO) is an essential component in environmental perception for autonomous driving. While 3D detectors using surround-view cameras are just flourishing, there is a growing tendency of using different transformer-based methods to learn queries in 3D space from 2D feature maps of perspective view. This paper proposes Sparse R-CNN 3D (SRCN3D), a novel two-stage fully-convolutional mapping pipeline for surround-view camera detection and tracking. SRCN3D adopts a cascade structure with twin-track update of both fixed number of proposal boxes and proposal latent features. Proposal boxes are projected to perspective view so as to aggregate Region of Interest (RoI) local features. Based on that, proposal features are refined via a dynamic instance interactive head, which then generates classification and the offsets applied to original bounding boxes. Compared to prior arts, our sparse feature sampling module only utilizes local 2D features for adjustment of each corresponding 3D proposal box, leading to a complete sparse paradigm. The proposal features and appearance features are both taken in data association process in a multi-hypotheses 3D multi-object tracking approach. Extensive experiments on nuScenes dataset demonstrate the effectiveness of our proposed SRCN3D detector and tracker. Code is available at https://github.com/synsin0/SRCN3D.
In the art of video editing, sound is really half the story. A skilled video editor overlays sounds, such as effects and ambients, over footage to add character to an object or immerse the viewer within a space. However, through formative interviews with professional video editors, we found that this process can be extremely tedious and time-consuming. We introduce Soundify, a system that matches sound effects to video. By leveraging labeled, studio-quality sound effects libraries and extending CLIP, a neural network with impressive zero-shot image classification capabilities, into a "zero-shot detector", we are able to produce high-quality results without resource-intensive correspondence learning or audio generation. We encourage you to have a look at, or better yet, have a listen to the results at https://chuanenlin.com/soundify.
Embedding models have been an effective learning paradigm for high-dimensional data. However, one open issue of embedding models is that their representations (latent factors) often result in large parameter space. We observe that existing distributed training frameworks face a scalability issue of embedding models since updating and retrieving the shared embedding parameters from servers usually dominates the training cycle. In this paper, we propose HET, a new system framework that significantly improves the scalability of huge embedding model training. We embrace skewed popularity distributions of embeddings as a performance opportunity and leverage it to address the communication bottleneck with an embedding cache. To ensure consistency across the caches, we incorporate a new consistency model into HET design, which provides fine-grained consistency guarantees on a per-embedding basis. Compared to previous work that only allows staleness for read operations, HET also utilizes staleness for write operations. Evaluations on six representative tasks show that HET achieves up to 88% embedding communication reductions and up to 20.68x performance speedup over the state-of-the-art baselines.