DeepInteraction++: Multi-Modality Interaction for Autonomous Driving

人工智能 模态(人机交互) 计算机科学 计算机视觉 模式识别(心理学)
作者
Zeyu Yang,Nan Song,Wei Li,Xiatian Zhu,Li Zhang,Philip H. S. Torr
出处
期刊:IEEE Transactions on Pattern Analysis and Machine Intelligence [IEEE Computer Society]
卷期号:: 1-14
标识
DOI:10.1109/tpami.2025.3565194
摘要

Existing top-performance autonomous driving systems typically rely on the multi-modal fusion strategy for reliable scene understanding. This design is however fundamentally restricted due to overlooking the modality-specific strengths and finally hampering the model performance. To address this limitation, in this work, we introduce a novel modality interaction strategy that allows individual per-modality representations to be learned and maintained throughout, enabling their unique characteristics to be exploited during the whole perception pipeline. To demonstrate the effectiveness of the proposed strategy, we design DeepInteraction++, a multi-modal interaction framework characterized by a multi-modal representational interaction encoder and a multi-modal predictive interaction decoder. Specifically, the encoder is implemented as a dual-stream Transformer with specialized attention operation for information exchange and integration between separate modality-specific representations. Our multi-modal representational learning incorporates both object-centric, precise sampling-based feature alignment and global dense information spreading, essential for the more challenging planning task. The decoder is designed to iteratively refine the predictions by alternately aggregating information from separate representations in a unified modality-agnostic manner, realizing multi-modal predictive interaction. Extensive experiments demonstrate the superior performance of the proposed framework on both 3D object detection and end-to-end autonomous driving tasks. Our code is available at https://github.com/fudan-zvg/DeepInteraction.

科研通智能强力驱动
Strongly Powered by AbleSci AI
科研通是完全免费的文献互助平台,具备全网最快的应助速度,最高的求助完成率。 对每一个文献求助,科研通都将尽心尽力,给求助人一个满意的交代。
实时播报
大卫完成签到,获得积分10
刚刚
沉静问芙完成签到 ,获得积分10
1秒前
陶醉的蜜蜂完成签到 ,获得积分10
3秒前
手术刀完成签到 ,获得积分10
5秒前
黑猫小苍完成签到,获得积分10
10秒前
柳crystal完成签到,获得积分10
11秒前
5易6完成签到 ,获得积分10
11秒前
15秒前
陈夏萍完成签到 ,获得积分10
17秒前
赵雪发布了新的文献求助10
19秒前
大模型应助一人采纳,获得10
19秒前
康轲完成签到,获得积分0
21秒前
赘婿应助核桃采纳,获得10
21秒前
领导范儿应助憨批采纳,获得10
24秒前
25秒前
爱学习的向日葵完成签到,获得积分10
25秒前
26秒前
28秒前
鲸落完成签到 ,获得积分10
28秒前
蚂蚁飞飞完成签到,获得积分10
28秒前
一人发布了新的文献求助10
31秒前
35秒前
郭子啊完成签到 ,获得积分10
36秒前
靳雪莲完成签到 ,获得积分10
36秒前
DKX完成签到 ,获得积分10
37秒前
夏洳完成签到 ,获得积分10
38秒前
憨批发布了新的文献求助10
39秒前
延娜完成签到,获得积分10
39秒前
光亮的盼完成签到 ,获得积分10
40秒前
40秒前
苑世朝完成签到,获得积分10
47秒前
51秒前
李子啊完成签到 ,获得积分10
52秒前
53秒前
摆哥完成签到,获得积分10
56秒前
guoxingliu完成签到,获得积分10
56秒前
YF发布了新的文献求助30
57秒前
上官若男应助科研通管家采纳,获得10
58秒前
58秒前
胡萝卜须完成签到 ,获得积分10
1分钟前
高分求助中
Principles of Economics, 11th Edition 10000
University Physics with Modern Physics, 16th edition 10000
(应助此贴封号)【重要!!请各用户(尤其是新用户)详细阅读】【科研通的精品贴汇总】 10000
Molecular Mechanisms of Photosynthesis, 4th Edition 1000
Organic Reactions, Volume 116 1000
Matrix Methods in Data Mining and Pattern Recognition 510
Social Skills Improvement System-Rating Scales--Chinese Version 500
热门求助领域 (近24小时)
化学 材料科学 医学 生物 纳米技术 工程类 有机化学 化学工程 生物化学 计算机科学 内科学 物理 复合材料 催化作用 细胞生物学 无机化学 光电子学 物理化学 电极 基因
热门帖子
关注 科研通微信公众号,转发送积分 7252936
求助须知:如何正确求助?哪些是违规求助? 8875060
关于积分的说明 18734667
捐赠科研通 6933491
什么是DOI,文献DOI怎么找? 3199831
关于科研通互助平台的介绍 2374606
邀请新用户注册赠送积分活动 2174506