Predicting the full appearance of a partially visible object, known as amodal content completion, remains a
challenging task. Many existing approaches struggle to recover missing regions without leaving artifacts from
the occluding objects or generating out-of-context content. To address this limitation, we introduce a two-stage
framework. In the first stage, a weighted mask is integrated with multiple layers of enhanced gated convolution
to generate an initial reconstruction while preserving mask awareness throughout feature extraction and
emphasizing the visible regions of the target object. In the second stage, this initial reconstruction is refined
by combining global contextual information with the extracted features through contextual attention and
skip connections, allowing the model to preserve spatial details and produce more coherent completions. We
evaluate our approach on the COCOA dataset and two of its subsets. The experimental results demonstrate
that the enhanced gated convolution and the contextual attention with skip connections improve reconstruction
quality, while the weighted mask provides complementary guidance when sufficient visible object information
is available. Together, these components produce reconstructions with richer semantic details and more
contextually accurate content than the baseline models, particularly for objects with uniform textures such
as animals.