CMDA: Cross-Modal and Domain Adversarial Adaptation for LiDAR-Based 3D Object Detection

Recent LiDAR-based 3D Object Detection (3DOD) methods show promising results, but they often do not generalize well to target domains outside the source (or training) data distribution. To reduce such domain gaps and thus to make 3DOD models more generalizable, we introduce a novel unsupervised domain adaptation (UDA) method, called CMDA, which (i) leverages visual semantic cues from an image modality (i.e., camera images) as an effective semantic bridge to close the domain gap in the cross-modal Bird's Eye View (BEV) representations. Further, (ii) we also introduce a self-training-based learning strategy, wherein a model is adversarially trained to generate domain-invariant features, which disrupt the discrimination of whether a feature instance comes from a source or an unseen target domain. Overall, our CMDA framework guides the 3DOD model to generate highly informative and domain-adaptive features for novel data distributions. In our extensive experiments with large-scale benchmarks, such as nuScenes, Waymo, and KITTI, those mentioned above provide significant performance gains for UDA tasks, achieving state-of-the-art performance.

ORA3D: Overlap Region Aware Multi-view 3D Object Detection

In multi-view 3D object detection tasks, disparity supervision over overlapping image regions substantially improves the overall detection performance. However, current multi-view 3D object detection methods often fail to detect objects in the overlap region properly, and the network's understanding of the scene is often limited to that of a monocular detection network. To mitigate this issue, we advocate for applying the traditional stereo disparity estimation method to obtain reliable disparity information for the overlap region. Given the disparity estimates as a supervision, we propose to regularize the network to fully utilize the geometric potential of binocular images, and improve the overall detection accuracy. Moreover, we propose to use an adversarial overlap region discriminator, which is trained to minimize the representational gap between non-overlap regions and overlapping regions where objects are often largely occluded or suffer from deformation due to camera distortion, causing a domain shift. We demonstrate the effectiveness of the proposed method with the large-scale multi-view 3D object detection benchmark, called nuScenes. Our experiment shows that our proposed method outperforms the current state-of-the-art methods.

Sound-Guided Semantic Image Manipulation

The recent success of the generative model shows that leveraging the multi-modal embedding space can manipulate an image using text information. However, manipulating an image with other sources rather than text, such as sound, is not easy due to the dynamic characteristics of the sources. Especially, sound can convey vivid emotions and dynamic expressions of the real world. Here, we propose a framework that directly encodes sound into the multi-modal (image-text) embedding space and manipulates an image from the space. Our audio encoder is trained to produce a latent representation from an audio input, which is forced to be aligned with image and text representations in the multi-modal embedding space. We use a direct latent optimization method based on aligned embeddings for sound-guided image manipulation. We also show that our method can mix text and audio modalities, which enrich the variety of the image modification. We verify the effectiveness of our sound-guided image manipulation quantitatively and qualitatively. We also show that our method can mix different modalities, i.e., text and audio, which enrich the variety of the image modification. The experiments on zero-shot audio classification and semantic-level image classification show that our proposed model outperforms other text and sound-guided state-of-the-art methods.