Nowadays, pre-training big models on large-scale datasets has become a crucial topic in deep learning. The pre-trained models with high representation ability and transferability achieve a great success and dominate many downstream tasks in natural language processing and 2D vision. However, it is non-trivial to promote such a pretraining-tuning paradigm to the 3D vision, given the limited training data that are relatively inconvenient to collect. In this paper, we provide a new perspective of leveraging pre-trained 2D knowledge in 3D domain to tackle this problem, tuning pre-trained image models with the novel Point-to-Pixel prompting for point cloud analysis at a minor parameter cost. Following the principle of prompting engineering, we transform point clouds into colorful images with geometry-preserved projection and geometry-aware coloring to adapt to pre-trained image models, whose weights are kept frozen during the end-to-end optimization of point cloud analysis tasks. We conduct extensive experiments to demonstrate that cooperating with our proposed Point-to-Pixel Prompting, better pre-trained image model will lead to consistently better performance in 3D vision. Enjoying prosperous development from image pre-training field, our method attains 89.3% accuracy on the hardest setting of ScanObjectNN, surpassing conventional point cloud models with much fewer trainable parameters. Our framework also exhibits very competitive performance on ModelNet classification and ShapeNet Part Segmentation. Code is available at https://github.com/wangzy22/P2P
Recent progress in vision Transformers exhibits great success in various tasks driven by the new spatial modeling mechanism based on dot-product self-attention. In this paper, we show that the key ingredients behind the vision Transformers, namely input-adaptive, long-range and high-order spatial interactions, can also be efficiently implemented with a convolution-based framework. We present the Recursive Gated Convolution ($\textit{g}^\textit{n}$Conv) that performs high-order spatial interactions with gated convolutions and recursive designs. The new operation is highly flexible and customizable, which is compatible with various variants of convolution and extends the two-order interactions in self-attention to arbitrary orders without introducing significant extra computation. $\textit{g}^\textit{n}$Conv can serve as a plug-and-play module to improve various vision Transformers and convolution-based models. Based on the operation, we construct a new family of generic vision backbones named HorNet. Extensive experiments on ImageNet classification, COCO object detection and ADE20K semantic segmentation show HorNet outperform Swin Transformers and ConvNeXt by a significant margin with similar overall architecture and training configurations. HorNet also shows favorable scalability to more training data and a larger model size. Apart from the effectiveness in visual encoders, we also show $\textit{g}^\textit{n}$Conv can be applied to task-specific decoders and consistently improve dense prediction performance with less computation. Our results demonstrate that $\textit{g}^\textit{n}$Conv can be a new basic module for visual modeling that effectively combines the merits of both vision Transformers and CNNs. Code is available at https://github.com/raoyongming/HorNet
Lookup table (LUT) has shown its efficacy in low-level vision tasks due to the valuable characteristics of low computational cost and hardware independence. However, recent attempts to address the problem of single image super-resolution (SISR) with lookup tables are highly constrained by the small receptive field size. Besides, their frameworks of single-layer lookup tables limit the extension and generalization capacities of the model. In this paper, we propose a framework of series-parallel lookup tables (SPLUT) to alleviate the above issues and achieve efficient image super-resolution. On the one hand, we cascade multiple lookup tables to enlarge the receptive field of each extracted feature vector. On the other hand, we propose a parallel network which includes two branches of cascaded lookup tables which process different components of the input low-resolution images. By doing so, the two branches collaborate with each other and compensate for the precision loss of discretizing input pixels when establishing lookup tables. Compared to previous lookup table-based methods, our framework has stronger representation abilities with more flexible architectures. Furthermore, we no longer need interpolation methods which introduce redundant computations so that our method can achieve faster inference speed. Extensive experimental results on five popular benchmark datasets show that our method obtains superior SISR performance in a more efficient way. The code is available at https://github.com/zhjy2016/SPLUT.
Grasping in dense clutter is a fundamental skill for autonomous robots. However, the crowdedness and occlusions in the cluttered scenario cause significant difficulties to generate valid grasp poses without collisions, which results in low efficiency and high failure rates. To address these, we present a generic framework called GE-Grasp for robotic motion planning in dense clutter, where we leverage diverse action primitives for occluded object removal and present the generator-evaluator architecture to avoid spatial collisions. Therefore, our GE-Grasp is capable of grasping objects in dense clutter efficiently with promising success rates. Specifically, we define three action primitives: target-oriented grasping for target capturing, pushing, and nontarget-oriented grasping to reduce the crowdedness and occlusions. The generators effectively provide various action candidates referring to the spatial information. Meanwhile, the evaluators assess the selected action primitive candidates, where the optimal action is implemented by the robot. Extensive experiments in simulated and real-world environments show that our approach outperforms the state-of-the-art methods of grasping in clutter with respect to motion efficiency and success rates. Moreover, we achieve comparable performance in the real world as that in the simulation environment, which indicates the strong generalization ability of our GE-Grasp. Supplementary material is available at: https://github.com/CaptainWuDaoKou/GE-Grasp.
Talking head synthesis is an emerging technology with wide applications in film dubbing, virtual avatars and online education. Recent NeRF-based methods generate more natural talking videos, as they better capture the 3D structural information of faces. However, a specific model needs to be trained for each identity with a large dataset. In this paper, we propose Dynamic Facial Radiance Fields (DFRF) for few-shot talking head synthesis, which can rapidly generalize to an unseen identity with few training data. Different from the existing NeRF-based methods which directly encode the 3D geometry and appearance of a specific person into the network, our DFRF conditions face radiance field on 2D appearance images to learn the face prior. Thus the facial radiance field can be flexibly adjusted to the new identity with few reference images. Additionally, for better modeling of the facial deformations, we propose a differentiable face warping module conditioned on audio signals to deform all reference images to the query space. Extensive experiments show that with only tens of seconds of training clip available, our proposed DFRF can synthesize natural and high-quality audio-driven talking head videos for novel identities with only 40k iterations. We highly recommend readers view our supplementary video for intuitive comparisons. Code is available in https://sstzal.github.io/DFRF/.
Objects are usually associated with multiple attributes, and these attributes often exhibit high correlations. Modeling complex relationships between attributes poses a great challenge for multi-attribute learning. This paper proposes a simple yet generic framework named Label2Label to exploit the complex attribute correlations. Label2Label is the first attempt for multi-attribute prediction from the perspective of language modeling. Specifically, it treats each attribute label as a "word" describing the sample. As each sample is annotated with multiple attribute labels, these "words" will naturally form an unordered but meaningful "sentence", which depicts the semantic information of the corresponding sample. Inspired by the remarkable success of pre-training language models in NLP, Label2Label introduces an image-conditioned masked language model, which randomly masks some of the "word" tokens from the label "sentence" and aims to recover them based on the masked "sentence" and the context conveyed by image features. Our intuition is that the instance-wise attribute relations are well grasped if the neural net can infer the missing attributes based on the context and the remaining attribute hints. Label2Label is conceptually simple and empirically powerful. Without incorporating task-specific prior knowledge and highly specialized network designs, our approach achieves state-of-the-art results on three different multi-attribute learning tasks, compared to highly customized domain-specific methods. Code is available at https://github.com/Li-Wanhua/Label2Label.
Rapid progress and superior performance have been achieved for skeleton-based action recognition recently. In this article, we investigate this problem under a cross-dataset setting, which is a new, pragmatic, and challenging task in real-world scenarios. Following the unsupervised domain adaptation (UDA) paradigm, the action labels are only available on a source dataset, but unavailable on a target dataset in the training stage. Different from the conventional adversarial learning-based approaches for UDA, we utilize a self-supervision scheme to reduce the domain shift between two skeleton-based action datasets. Our inspiration is drawn from Cubism, an art genre from the early 20th century, which breaks and reassembles the objects to convey a greater context. By segmenting and permuting temporal segments or human body parts, we design two self-supervised learning classification tasks to explore the temporal and spatial dependency of a skeleton-based action and improve the generalization ability of the model. We conduct experiments on six datasets for skeleton-based action recognition, including three large-scale datasets (NTU RGB+D, PKU-MMD, and Kinetics) where new cross-dataset settings and benchmarks are established. Extensive results demonstrate that our method outperforms state-of-the-art approaches. The source codes of our model and all the compared methods are available at https://github.com/shanice-l/st-cubism.
Most state-of-the-art instance-level human parsing models adopt two-stage anchor-based detectors and, therefore, cannot avoid the heuristic anchor box design and the lack of analysis on a pixel level. To address these two issues, we have designed an instance-level human parsing network which is anchor-free and solvable on a pixel level. It consists of two simple sub-networks: an anchor-free detection head for bounding box predictions and an edge-guided parsing head for human segmentation. The anchor-free detector head inherits the pixel-like merits and effectively avoids the sensitivity of hyper-parameters as proved in object detection applications. By introducing the part-aware boundary clue, the edge-guided parsing head is capable to distinguish adjacent human parts from among each other up to 58 parts in a single human instance, even overlapping instances. Meanwhile, a refinement head integrating box-level score and part-level parsing quality is exploited to improve the quality of the parsing results. Experiments on two multiple human parsing datasets (i.e., CIHP and LV-MHP-v2.0) and one video instance-level human parsing dataset (i.e., VIP) show that our method achieves the best global-level and instance-level performance over state-of-the-art one-stage top-down alternatives.
Different people age in different ways. Learning a personalized age estimator for each person is a promising direction for age estimation given that it better models the personalization of aging processes. However, most existing personalized methods suffer from the lack of large-scale datasets due to the high-level requirements: identity labels and enough samples for each person to form a long-term aging pattern. In this paper, we aim to learn personalized age estimators without the above requirements and propose a meta-learning method named MetaAge for age estimation. Unlike most existing personalized methods that learn the parameters of a personalized estimator for each person in the training set, our method learns the mapping from identity information to age estimator parameters. Specifically, we introduce a personalized estimator meta-learner, which takes identity features as the input and outputs the parameters of customized estimators. In this way, our method learns the meta knowledge without the above requirements and seamlessly transfers the learned meta knowledge to the test set, which enables us to leverage the existing large-scale age datasets without any additional annotations. Extensive experimental results on three benchmark datasets including MORPH II, ChaLearn LAP 2015 and ChaLearn LAP 2016 databases demonstrate that our MetaAge significantly boosts the performance of existing personalized methods and outperforms the state-of-the-art approaches.
In this paper, we present a new approach for model acceleration by exploiting spatial sparsity in visual data. We observe that the final prediction in vision Transformers is only based on a subset of the most informative tokens, which is sufficient for accurate image recognition. Based on this observation, we propose a dynamic token sparsification framework to prune redundant tokens progressively and dynamically based on the input to accelerate vision Transformers. Specifically, we devise a lightweight prediction module to estimate the importance score of each token given the current features. The module is added to different layers to prune redundant tokens hierarchically. While the framework is inspired by our observation of the sparse attention in vision Transformers, we find the idea of adaptive and asymmetric computation can be a general solution for accelerating various architectures. We extend our method to hierarchical models including CNNs and hierarchical vision Transformers as well as more complex dense prediction tasks that require structured feature maps by formulating a more generic dynamic spatial sparsification framework with progressive sparsification and asymmetric computation for different spatial locations. By applying lightweight fast paths to less informative features and using more expressive slow paths to more important locations, we can maintain the structure of feature maps while significantly reducing the overall computations. Extensive experiments demonstrate the effectiveness of our framework on various modern architectures and different visual recognition tasks. Our results clearly demonstrate that dynamic spatial sparsification offers a new and more effective dimension for model acceleration. Code is available at https://github.com/raoyongming/DynamicViT