Compressing Large Language Models by Streamlining the Unimportant Layer

Large language models (LLM) have been extensively applied in various natural language tasks and domains, but their applicability is constrained by the large number of parameters of the models. Consequently, there is an increasing emphasis on compact models that exhibit high performance. In this study, we observe that different layers in LLM have varying degrees of perturbation on the hidden states, which allows us to identify less important layers. Based on this phenomenon, we propose LLM-Streamline, which consists of two parts: layer pruning, where we remove a set of consecutive layers with the lowest importance in the model according to the target sparsity; and layer replacement, where we train a lightweight model to substitute the pruned layers, thereby mitigating the performance degradation caused by pruning. In our experiments, we utilize structures such as a multi-layer perceptron (MLP) and a transformer layer as lightweight models and ultimately demonstrate that a single MLP can effectively fit the pruned layers. Comprehensive experiments show that our proposed method, LLM-Streamline, outperforms previous state-of-the-art (SOTA) model pruning methods.

Direct-a-Video: Customized Video Generation with User-Directed Camera Movement and Object Motion

Recent text-to-video diffusion models have achieved impressive progress. In practice, users often desire the ability to control object motion and camera movement independently for customized video creation. However, current methods lack the focus on separately controlling object motion and camera movement in a decoupled manner, which limits the controllability and flexibility of text-to-video models. In this paper, we introduce Direct-a-Video, a system that allows users to independently specify motions for one or multiple objects and/or camera movements, as if directing a video. We propose a simple yet effective strategy for the decoupled control of object motion and camera movement. Object motion is controlled through spatial cross-attention modulation using the model's inherent priors, requiring no additional optimization. For camera movement, we introduce new temporal cross-attention layers to interpret quantitative camera movement parameters. We further employ an augmentation-based approach to train these layers in a self-supervised manner on a small-scale dataset, eliminating the need for explicit motion annotation. Both components operate independently, allowing individual or combined control, and can generalize to open-domain scenarios. Extensive experiments demonstrate the superiority and effectiveness of our method. Project page: https://direct-a-video.github.io/.

Measuring the Discrepancy between 3D Geometric Models using Directional Distance Fields

Qualifying the discrepancy between 3D geometric models, which could be represented with either point clouds or triangle meshes, is a pivotal issue with board applications. Existing methods mainly focus on directly establishing the correspondence between two models and then aggregating point-wise distance between corresponding points, resulting in them being either inefficient or ineffective. In this paper, we propose DirDist, an efficient, effective, robust, and differentiable distance metric for 3D geometry data. Specifically, we construct DirDist based on the proposed implicit representation of 3D models, namely directional distance field (DDF), which defines the directional distances of 3D points to a model to capture its local surface geometry. We then transfer the discrepancy between two 3D geometric models as the discrepancy between their DDFs defined on an identical domain, naturally establishing model correspondence. To demonstrate the advantage of our DirDist, we explore various distance metric-driven 3D geometric modeling tasks, including template surface fitting, rigid registration, non-rigid registration, scene flow estimation and human pose optimization. Extensive experiments show that our DirDist achieves significantly higher accuracy under all tasks. As a generic distance metric, DirDist has the potential to advance the field of 3D geometric modeling. The source code is available at \url{https://github.com/rsy6318/DirDist}.

Uni-paint: A Unified Framework for Multimodal Image Inpainting with Pretrained Diffusion Model

Recently, text-to-image denoising diffusion probabilistic models (DDPMs) have demonstrated impressive image generation capabilities and have also been successfully applied to image inpainting. However, in practice, users often require more control over the inpainting process beyond textual guidance, especially when they want to composite objects with customized appearance, color, shape, and layout. Unfortunately, existing diffusion-based inpainting methods are limited to single-modal guidance and require task-specific training, hindering their cross-modal scalability. To address these limitations, we propose Uni-paint, a unified framework for multimodal inpainting that offers various modes of guidance, including unconditional, text-driven, stroke-driven, exemplar-driven inpainting, as well as a combination of these modes. Furthermore, our Uni-paint is based on pretrained Stable Diffusion and does not require task-specific training on specific datasets, enabling few-shot generalizability to customized images. We have conducted extensive qualitative and quantitative evaluations that show our approach achieves comparable results to existing single-modal methods while offering multimodal inpainting capabilities not available in other methods. Code will be available at https://github.com/ysy31415/unipaint.

Enhancing Dynamic Image Advertising with Vision-Language Pre-training

In the multimedia era, image is an effective medium in search advertising. Dynamic Image Advertising (DIA), a system that matches queries with ad images and generates multimodal ads, is introduced to improve user experience and ad revenue. The core of DIA is a query-image matching module performing ad image retrieval and relevance modeling. Current query-image matching suffers from limited and inconsistent data, and insufficient cross-modal interaction. Also, the separate optimization of retrieval and relevance models affects overall performance. To address this issue, we propose a vision-language framework consisting of two parts. First, we train a base model on large-scale image-text pairs to learn general multimodal representation. Then, we fine-tune the base model on advertising business data, unifying relevance modeling and retrieval through multi-objective learning. Our framework has been implemented in Baidu search advertising system "Phoneix Nest". Online evaluation shows that it improves cost per mille (CPM) and click-through rate (CTR) by 1.04% and 1.865%.

Resilient conductive membrane synthesized by in-situ polymerisation for wearable non-invasive electronics on moving appendages of cyborg insect

By leveraging their high mobility and small size, insects have been combined with microcontrollers to build up cyborg insects for various practical applications. Unfortunately, all current cyborg insects rely on implanted electrodes to control their movement, which causes irreversible damage to their organs and muscles. Here, we develop a non-invasive method for cyborg insects to address above issues, using a conformal electrode with an in-situ polymerized ion-conducting layer and an electron-conducting layer. The neural and locomotion responses to the electrical inductions verify the efficient communication between insects and controllers by the non-invasive method. The precise "S" line following of the cyborg insect further demonstrates its potential in practical navigation. The conformal non-invasive electrodes keep the intactness of the insects used while controlling their motion. With the antennae, important olfactory organs of insects preserved, the cyborg insect, in the future, may be endowed with abilities to detect the surrounding environment.

GeoUDF: Surface Reconstruction from 3D Point Clouds via Geometry-guided Distance Representation

The recent neural implicit representation-based methods have greatly advanced the state of the art for solving the long-standing and challenging problem of reconstructing a discrete surface from a sparse point cloud. These methods generally learn either a binary occupancy or signed/unsigned distance field (SDF/UDF) as surface representation. However, all the existing SDF/UDF-based methods use neural networks to implicitly regress the distance in a purely data-driven manner, thus limiting the accuracy and generalizability to some extent. In contrast, we propose the first geometry-guided method for UDF and its gradient estimation that explicitly formulates the unsigned distance of a query point as the learnable affine averaging of its distances to the tangent planes of neighbouring points. Besides, we model the local geometric structure of the input point clouds by explicitly learning a quadratic polynomial for each point. This not only facilitates upsampling the input sparse point cloud but also naturally induces unoriented normal, which further augments UDF estimation. Finally, to extract triangle meshes from the predicted UDF we propose a customized edge-based marching cube module. We conduct extensive experiments and ablation studies to demonstrate the significant advantages of our method over state-of-the-art methods in terms of reconstruction accuracy, efficiency, and generalizability. The source code is publicly available at https://github.com/rsy6318/GeoUDF.

CorrI2P: Deep Image-to-Point Cloud Registration via Dense Correspondence

Motivated by the intuition that the critical step of localizing a 2D image in the corresponding 3D point cloud is establishing 2D-3D correspondence between them, we propose the first feature-based dense correspondence framework for addressing the image-to-point cloud registration problem, dubbed CorrI2P, which consists of three modules, i.e., feature embedding, symmetric overlapping region detection, and pose estimation through the established correspondence. Specifically, given a pair of a 2D image and a 3D point cloud, we first transform them into high-dimensional feature space and feed the resulting features into a symmetric overlapping region detector to determine the region where the image and point cloud overlap each other. Then we use the features of the overlapping regions to establish the 2D-3D correspondence before running EPnP within RANSAC to estimate the camera's pose. Experimental results on KITTI and NuScenes datasets show that our CorrI2P outperforms state-of-the-art image-to-point cloud registration methods significantly. We will make the code publicly available.

Part-level Action Parsing via a Pose-guided Coarse-to-Fine Framework

Action recognition from videos, i.e., classifying a video into one of the pre-defined action types, has been a popular topic in the communities of artificial intelligence, multimedia, and signal processing. However, existing methods usually consider an input video as a whole and learn models, e.g., Convolutional Neural Networks (CNNs), with coarse video-level class labels. These methods can only output an action class for the video, but cannot provide fine-grained and explainable cues to answer why the video shows a specific action. Therefore, researchers start to focus on a new task, Part-level Action Parsing (PAP), which aims to not only predict the video-level action but also recognize the frame-level fine-grained actions or interactions of body parts for each person in the video. To this end, we propose a coarse-to-fine framework for this challenging task. In particular, our framework first predicts the video-level class of the input video, then localizes the body parts and predicts the part-level action. Moreover, to balance the accuracy and computation in part-level action parsing, we propose to recognize the part-level actions by segment-level features. Furthermore, to overcome the ambiguity of body parts, we propose a pose-guided positional embedding method to accurately localize body parts. Through comprehensive experiments on a large-scale dataset, i.e., Kinetics-TPS, our framework achieves state-of-the-art performance and outperforms existing methods over a 31.10% ROC score.

A Baseline Framework for Part-level Action Parsing and Action Recognition

This technical report introduces our 2nd place solution to Kinetics-TPS Track on Part-level Action Parsing in ICCV DeeperAction Workshop 2021. Our entry is mainly based on YOLOF for instance and part detection, HRNet for human pose estimation, and CSN for video-level action recognition and frame-level part state parsing. We describe technical details for the Kinetics-TPS dataset, together with some experimental results. In the competition, we achieved 61.37% mAP on the test set of Kinetics-TPS.